Islands in the Sky

Citation preview

Islands in the Sky

Islands in the Sky Bold New Ideas for Colonizing Space

Edited by

Stanley Schmidt and Robert Zubrin

WILEYPOPULARSCIENCE

John Wiley & Sons, Inc. New York • Chichester• Brisbane • Toronto • Singapore

This text is printed on acid-free paper. Copyright © 1996 Bantam Doubleday Dell Magazines Published by John Wiley & Sons, Inc. All rights reserved. Published simultaneously in Canada. Reproduction or translation of any part of this work beyond chat permitted by Section 107 or 108of the 1976 United Scates Copyright Act without the permissionof the copyright owner is unlawful. Requests for permission or further information should be addressed to the Permissions Department, John Wiley & Sons, Inc. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. le is sold with the understanding that the publisher is not engaged in rendering legal, accounting, or other professional services. If legal advice or other expert assistance is required, the services of a competent professional person should be sought.

Libraryof CongressCataloging-in-Publication Data Reaching for the Stars : Analog's book of space exploration/ edited by Robert Zubrin and Stanley Schmidt. p. cm. ISBN 0-471-13561-5 (pbk : alk. paper) I. Outer space--Exploration. 2. Space colonies. I. Zubrin, Robert. 11. Schmidt, Stanley. QB500.R33 1996 95-32914 9l9.9'04--dc20

Printed in the United States of America

10 9 8 7 6 5 4 3 2 1

Contents

1

Introduction Part One

Breaking the Bonds of the Earth

I

Comes the Revolution .. G. Harry Stine

7

2 The Hypersonic Skyhook Robert M. Zubrin

27

Part Two

Stepping Into the Solar System

3

Mars Direct: A Proposal for the Rapid Exploration and Colonization of the Red Planet Robert M. Zubrin and David A. Baker

43

4

Inward Ho! Stephen L. Gillett

71

5 Colonizing the Outer Solar System Robert M. Zubrin V

85

vi

•

Contents

6 Islands in the Sky: Human Exploration

95

and Settlement of the Oort Cloud

RichardP. Terra

7 Alien Life Between Here and the Stars Dr. RobertL. Forward Part Three

117

Creating New Worlds

8 T erraforming Mars RobertM. ZubrinandChristopher P. McKay

125

9 A Planet Dweller's Dreams MartynJ. Fogg

147

10 Astrophysical Engineering and the Fate

169

of the Earth

MartynJ. Fogg Part Four

Advanced Drives and Interstellar Travel

11 To the Stars! GordonR. Woodcock

183

12 The Magnetic Sail

199

RobertM. Zubrin 13 The Tachyon Drive: Infinite Exhaust Velocity at Zero Energy Cost

221

JohnG. Cramer

14 The Negative Matter Space Drive

227

Dr. RobertL. Forward

15 The Economics of Interstellar Commerce

243

WarrenSalomon References

261

Introduction

At the 1989 International Space Development Conference in Chicago, one of Analog's writers advised me to be sure to hear a talk by someone named Robert Zubrin, on a scheme for greatly reducing the cost of traveling to other planets. It sounded at least casually interesting, but the program showed that the paper was being presented shortly after my departure. Since at that time I knew nothing about Bob Zubrin or his ideas, I simply shrugged off his talk. It was too bad I'd have to miss it, but there were plenty of others I'd also be missing, and I had no reason to assume this one would be any of course, that recommendation more of a loss than them-except, from our mutual friend. What I didn't know was that the mutual friend had also told Bob about me. Late one afternoon, as I was passing through the hotel lobby on my way to something utterly unrelated, a visibly energetic young man accosted me and said, "Dr. Schmidt?" "Yes," I confessed cautiously. "l 'm Bob Zubrin, 11 he said. "l 'm giving a paper tomorrow that I think you might be interested in for Analog, and I hear you won't be " able to be there. I wondered if you might have a few minutes .... My normal reaction to such encounters is to cringe and start thinking about how to escape if, as often happens, the unsolicited exposition proves long, tedious, and fruitless. This time, though, something told me to listen, though I must confess I still started out with a bit of the usual skepticism. 1

2

•

Islands in the Sky

That didn't last long. Bob Zubrin proceeded to give me, there in the lobby, what amounted to a private performance of the public talk he was going to give the next day. He gave it much faster than I suspect he did in public, perhaps because he sensed that l 'd been on my way to something. ( It may be significant that I no longer remember what.) He didn't refer to notes, except to point something out in a diagram or table. He never missed a beat and always had a ready answer if I had a question. l 've seldom seen a speaker so thoroughly familiar with what he had to say. And what he had to say was spellbinding. Why should spacecrafts spend much of their fuel lugging all the fuel for a round trip, when it's perfectly possible to refuel at your destination? It was the kind of idea that makes you say, "Why didn't I think of that?" and Bob made a detailed, thoroughly thought-out, compelling case for it. As l 've already hinted, my usual reaction when somebody collars me in a hallway to tell me about a story or article they want to write, is either to tell them why we can't use it or to say cautiously, "Well, if you want to send it in, I'll read it. 11 This time I found myself saying, "How soon can I have the article ?11 Only after we had gone our separate ways did I find the time and presence of mind to think, "What have I done?" But the article came in right when Bob had promised, and in such polished form that it needed very little editorial work. We published it, and our annual poll showed it to be one of our readers' favorites for the year. I have since come to regard Bob Zubrin as one of the most prolific and reliable sources of solid, imaginative ideas for space travel-an opinion widely shared in the aerospace field. Now, whenever he calls me to describe a new idea, I drop what I'm doing and listen. Usually I buy the article; you'll find several of them in this book. Why are such ideas important? Humankind needs to get into space, for a wide variety of reasons. The most basic is survival itself. We know from the paleontological record that mass extinctions have occurred on this planet in the past. We know that they will almost certainly happen again; they could happen at any time, and we could be the subject of one. But disasters affecting a whole planet occur much oftener than disasters affecting an entire solar system or galaxy. We as a species will be far less vulnerable if and when we have outposts elsewhere. Beyond that, there's the fact that Earth's human population is growing at such a rate that it's beginning to feel the pinch of limited

Introduction

•

3

elbow room and resources. This has several deleterious effects: the direct ones of trying to house and feed billions of people with fixed resources, and the indirect ones of the psychological impact of feeling that there's no more room to grow. Easy access to space can do something about both kinds of problems. It won't eliminate the need for people to exercise reasonable prudence in reproducing and using whatever space and resources they have. But it will provide more space for people to move into, more material resources for people both down here and out there, and the psychological and cultural boost that comes from the existence of a new frontier. The key word there is easy. Access to space has seldom been seen as easy; indeed, it's often been used as a metaphor for the ultimate in difficulty. ("If we can put a man on the Moon ... ") But things often seem prohibitively difficult not because they are, but simply because we haven't yet learned to do them-and often we must learn the hard ways before we can see the easy ones. When I was growing up, many adults seriously believed that space travel of any kind was quite impossible. That was true even less than five years before Sputnik and less than twenty before Apollo 11. Before that, people were just as skeptical about air travel, which is now routine on a very large scale. Not so long ago, the very concept of the telephone seemed like the wildest fantasy. A little later, people knew it could be done, but only the truly visionary dreamed of a time when every village would have a phone. Now, in many parts of the world, not only does every home have at least one phone, but many people have phones that they can use anywhere-even at the beach. So when somebody like Bob Zubrin tells you that a ticket to space may someday be as cheap and easy to get as a ticket to Chicago is now, there's a good chance he's right. How do you get from "impossible" to "routine"? You need people who can look beyond what "Everyone knows" and see what we could do in the future. That's where Analog and the writers in this book come in. Analog Science Fiction & Fact was born in 1930 and has been published almost continuously since then. For its first thirty years its title was some version of Astounding Science Fiction, but science fact has always been an important part of its make-up, both as the basis for its fiction and as the subject of factual articles like those gathered here. Its stories and articles in past decades have played a demonstrably important role in getting us as far as we've come so far. Many scientists and engineers responsible for real

4

•

Islands in the Sky

achievements in space were originally drawn to their careers by the ideas they found here. Others saw them as wild imaginings. They thought, "I can make that happen"-and they did. Recently space exploration has fallen on hard times. It has lost popular support; too many people have lost sight of why we need it. But we still do need it; and to get the huge benefits it has to offer, we need people who can see, and show others, the huge range of possibilities, and how we can bring them within our grasp. Astounding in the past was a leading source of such people and ideas. Thanks largely to them, we have made some preliminary accomplishments. But the real payoffs still lie ahead, and Analog today is still pointing the way to them. In this book you'll find a treasury of imaginative suggestions from some of the brightest minds working in this field. And remember: we're not kidding about any of this.1hese articles are not "just science fiction." They are things we can do-and with any luck at all, and vision and determination, we will. Stanley Schmidt Editor, Analog Science Fictionand Fact

Part One

Breaking the Bonds of the Earth

Chapter 1

G . Harry Stine

Comes the Revolution

• • • •

In -.pitl' lit° cra:-hes, politics, hudget hattles, and opposmon from tl1lLi\''s ,·l'rsinn of Juk·s Verne's Baltimore Gun Club, we're on the \'l•r.!.!L' llf ha,·ing inexpensive, reliable, and responsive single-stagetlH1rhir (SSTO) spaceships like the Delta Clipper DC-Y described in [)r. Arlan Andrews's science-fact article in the June 1993 issue. 1 C1in.'n rhat \\'L' can build reusable spaceships that operate like commercial airliners, what is this really going to mean? In one word: rl!1.olttticm. In the social context, the word "revolution" means a rapid and radical change in the way of doing things as opposed to "evolution" which is a slow and gradual change. And the introduction of the SSTO is indeed revolutionary, as you will see herein. It's presumptuous to believe that anyone can fully assess the impacts of a truly revolutionary concept such as the SSTO. However, assuming that the basic technology to build SSTO spaceships exists, it's possible to look at how SSTO spaceships will ( 1) reduce the costs of space transportation, (2) change the way we conduct operations in the Earth-Moon system, and (3) impact allied areas that haven't been considered when operating with expendable space launch vehicles. Although an SSTO spaceship hasn't flown yet, it's prudent to begin looking at the consequences of this space transportation revolu1

7

8

•

Islands in the Sky

tion. It is not too early to look at the business economics of the revolution. (The question was unreliably reported to have been asked in the Continental Congress, "Mister Chairman, as the delegate from New Hampshire, may I ask if this revolution is going to cost very much?") And given the lightning speed with which our political, regulatory, and legal systems operate, is it wise to leave political, regulatory, and legal questions unresolved until the last minute? We may end up having the technology for a cheap, reliable, and reusahle SSTO spaceship but be unable to use its services or even fly it!

THE TECHNOLOGYIS IMMATERIAL First off, assume that the basic technology to build SSTO spaceships exists. Leave it to others to focus on technology; they lm·e the technology. However, even after nearly two decades of peopk like me admonishing them, they still don't understand that technology alone isn't going to make it happen. Where specific technological problems presently exist, aviation history says that a rcasonahle number of flight tests with prototype X-vehicles can provide engineering solutions. Build 'em, boost 'em, and bend 'em ... and name a few more streets at Edwards Air Force Base for courageous former test pilots. Furthermore, the specific SSTO technology used is immaterial. For example, the exact operational modes-VTOVL, VTOHL, HTOHL, nose-first entry, tail-first entry, etc.-influence only the type of required spaceport facilities, not their extent. Basic assumptions about a generic SSTO include: • Propellants are liquid oxygen and liquid hydrogen with no auxiliary propellants. • The airframe, propulsion, avionics, and other on-board systems are completely reused. • Turnaround between flights is no longer than 72 hours. • A ground crew of 50 technicians and mechanics per flight is required. • Operations invol~e only aircraftlike procedures from austere "flight simplex" facilities.

Comes the Revolution

•

9

THE ECONOMY, STUPID! Listen up! The first lesson: The space transportation revolution will not occur unless someone can make money flying spaceships. For the last 30 years, space transportation has been operated and controlled by the federal government as a part of national security. However, with the end of the Cold War, it's too expensive to continue doing it the old way. And this socialized space transportation system has retarded progress. It has been 35 years since NASA was established. This is the same period that existed between the first government air mail flight with DeHaviland DH-9 biplanes in 1919 and the first flight of the Boeing 707 prototype in 1954. Anyone who claims that srace technology is more complex than aeronautical technology simply doesn't understand what went on between 1919 and 1954. The federal government has shown it cannot operate a spaceline any more than it could (or did) operate a national railroad or a national airline. Private enterprise did and does. With the SSTO, it's going to make more sense to let private enterprise engage in profitable space ventures. However, the bad news is: Perceivedrisk is high. The good news is: It has always been high! In the past, high-risk ventures with new technology have used a proven risk reduction procedure: modest government research funding and incentive programs to reduce risk. Railroads got government guaranteed loans and land grants. Airlines got airliners built using government research findings and air mail contracts to help them get started. The government is presently comtinuing this historic procedure by funding experimental SSTO test vehicles. As we have learned in the past 35 years, few things get done well and economically without the profit motive. Before we go any farther, we must determine the attractiveness of the business opportunity of providing civil and military space transportation. Therefore, we must first answer the question: Can someone make money doing it? The answer to this question rests with the size of the market opportunity. This is, in turn, dependent upon the nature and price of the service offered by spaceship operators. The market opportunity and the price of the service are interconnected by a feedback loop. One affects the other.

10 •

Islandsin the Sky

With this principle in mind, The Enterprise Institute, Inc., in Phoenix, Arizona, developed a space transportation economics model designed to answer the questions an investment broker or a banker would ask the organizers of a prototype spaceline who are raising capital to buy a fleet of SSTOs. 2 A banker or investor wants an answer to a simple question: Am I going to get my money back? In short, what's the risk?Can the service be sold? Is there a market? He will ask additional questions that will get him an answer to this first question.

WHATIS THE MARKET? The effect of introducing a new product or service into any market ranges from zero to revolutionary. An evaluation requires answers to several questions before the effect of a new product or market opportunity can be assessed. These include:

• WHAT-What is it and what will it do? For the purposes of this analysis, a first-generation SSTO spaceship will transport up to 10 tons of cargo and/or passengers into low-Earth orbit and back every 3 days.

• LIMITS-What won't it do?In terms of its limitations, a firstgeneration SSTO spaceship will not be configured to accommodate very large and bulky payloads. Payload support services for cargo will be limited as is the case with current cargo servicesprovided by airline operators. Customers must supply their own specializedcargo support. • AVAILABILITY-When, where,and how easy is the product or sen,,iceto obtain?Each SSTO will provide service at regular intervals of 3 days. A full payload bay won't be required for a flight. Service will be provided from several spaceports. Weather will be no more of a problem than it is for subsonic commercial air transportation. • PRICE-How muchwillit cost?This raises the key issue: How much will a commercialspaceshipoperator needto chargefor the transport service?How does this price compare to alternative transport services to low-Earth orbit?

Comes the Revolution . . . .

•

11

The requiredrevenue per fl.ightcan be determined if the following performance and financial information is known. 1. Capital equipment costs-Spaceship, ground support equipment and facilities, spaceship life, and facilities life. G&A, insurance, spaceship 2. Operating costs-Marketing, maintenance, crew and ground personnel, and fuel. 3. Financial costs-Competitive return on equity and cost of debt. 4. Capital structure of the business-Debt to equity ratio. 5. Performancedata-Weight profile, fleet attrition, and flight rates. The Enterprise Institute financial model was designed to allow variation in inputs to determine factors having the most significant effect upon the required revenue per flight. One other point about the financial model: It was assumed that, at the end of S years, the business was totally liquidated. The model shows no value for residual costs, market position, or used equipment. For a first-generation spaceship, the differences between a commercial vehicle and a military vehicle converge. Thus, if a first-generation spaceship is designed to satisfy military and civil customers-as was done with the Boeing 707-a manufacturer can spread his nonrecurring costs over a wider base. This in turn will reduce the cost of a spaceship and make it even more economically viable in the commercial sector. Furthermore, satisfying both types of customers through convergence not only lowers development costs but also reduces the risk involved in developing the technology. The bankers will like that. The most realistic run with the model used the following key assumptions: A spaceship costs $500 million ( 1993 dollars), has a life of 500 flights, requires 1,200,000 pounds of liquid oxygen and liquid hydrogen per flight at a cost of $0.50 per pound, and has an annual flight rate of 100 flights. In addition, a conservative debt to equity capital structure of 33% debt and 67% equity was assumed for the business. A competitive pretax return on equity requirement was pegged at 16%.

12

•

Islands in the Sky

The results are impressive.The total required revenue per flight for placing 10 tons of cargo and/or passengers into LEO is $1.6 million ( 1993dollars). In terms of passengers, this works out to be $14,500per person for a round,trip ticket to low,Earth orbit. ( If you want to go, you don't buy a new car that year.) This represents a price about 1 to 3% of current launch services. But is this reasonable? What assumptions exert heavy influence on the competitive posture of a reusable spaceship relative to expendable and semi,expendable vehicles? Various sensitivity analyses identified those factors that have the greatest effect upon the required revenue per flight-that is, they identified the economic "drivers." Financial model inputs of particular interest were spaceship pur, chase price, propellant cost, spaceship life, and number of flights per year. The price of anythingvaries according to the number of produc, tion units over which development costs will be recovered, the recurringcost of production, and the associated production learning curve. The cost of propellants depends on the overall demand for them. The flight rate is determined by spaceship design. Spaceship designsthat have a longer life and allow faster turnaround influence the revenue needed to provide an attractive return on investment. The results of the sensitivity analysis are interesting: • A five,fold increase in the price of the propellants from $0.50 to $2.50per pound caused the required revenue per flight to increase to $2.2 million from $1.6 million, a 38% increase. • When the propellant price of $0.50per pound was maintained, but the price of a spaceship was tripled from $500 million to $1.5 billion, the required revenue rose from $1.6 million to $4.2 million, a 163%increase. Of the two inputs, the purchase price of a spaceship clearly is the significant economic driver. It is important to note that in the case of a fuel price of $2.50 per pound and a spaceship cost of $1.Sbillion, the resulting required revenue of $4.7 million per flight remains a fraction of the cost of service using expendable and semi,expendable launch vehicles. Other sensitivity analysesshowed great flexibility between spaceship cost, spaceship life, and spaceship flight rates.

Comes the Revolution . . . .

•

13

For those who believe that certification costs could seriously skew the model's results, these factors were taken into account. The increased development costs simply increase the basic spaceship cost from, for example, $500 million to $700 million, or from $1.2 billion to $1.4 billion. This means that the prospective operators will have to do a little more work when talking to the bankers. The model was run with a spaceship cost of $2 billion. In such an example, the financial community is likely to conclude that the venture is slightly more risky. Therefore, they'll want it to be more heavily equity oriented and with a greater after tax return on equity of 20+ %. In this case, the required revenue became $10 million for 10 tons, still extremely competitive against existing expendable systems. The sensitivity analysis underscores what every airline in the world knows: Keep the airplaneflying. The primary economic driver is the cost of the equipment and the efficiency of its use. Operating costs such as fuel become more important only when all competitors are using vehicles of a similar nature. Although further financial analysis of spaceship characteristics and economics will define the exact nature of the costs, operating profile and resulting required revenue, the basic conclusion will still stand: A spaceshipwith performanceand cost characteristicsin the ballpark with those outlined will hold an overwhelmingprice advantage ouer any expendableor semi-expendablespacetransportationsystem. The flexibility of operation inherent in an SSTO spaceship will extend this competitive advantage still further. From an investor's point of view, the financial analysis shows that a space transportation service using SSTO spaceships can be brought to market at a price that can only be described as revolutionary. Lowering the price of a valuable service to between 1 and 5% of current prices is not an event that goes unnoticed in the business community.

MARKET IMPACTS AND CONSEQUENCES Whenever such a major change is brought to a market, market dynamics change to a degree that causes major ripples through other markets. A good example of this is the introduction of communica-

14 •

Islands in the Sky

tions satellites and their effect on the television industry, television programming,and long-distance and intercontinental telephone service. The rapid decrease in the cost of high capacity integrated circuits allowed the development of the personal computer that, in tum, created major changes in the computer industry as a whole and the manner in which companies, both large and small, conduct business. Similarly,the introduction of SSTO spaceships into commercial service will revolutionize the infant commercial space transportation industry. Regular and inexpensive transport services to low-Earth orbit will cause a major shift in the design and operating philosophy of satellites, space stations, space-based production facilities, Earthlunar travel and interplanetary travel. Present satellite design and operating philosophy is the result of three factors: (1) The high cost of getting to low-Earth orbit, (2) the lack of a significant human presence in space, and (3) the long and uncertain lead times in obtaining space transportation services. Reducing the cost of reaching low-Earth orbit has the greatest impact. Robert A. Heinlein said it 40 years ago, and we should have paid more attention to him. When the cost of climbing out of the Earth's gravity is substantially lowered, the heavy emphasis placed upon the weight and redundancy of satellites, space stations, anJ other space facilities will be reduced. The decreased emphasis on very high reliability required by one-shot operating philosophies permits the utilization of less exotic materials and the application of modular spacecraftdesign. These factors will result in less expensive satellites and on-orbit facilities and, in tum, an increase in market demand for the services. Low cost access to LEO will also allow more energy-intensiveflights for orbital, translunar, and interplanetary travel. A greater human presence in space will also have an effect upon satellite design. The ability to reach orbit easily and quickly permits on-orbit repair and lowers the need for heavily redundant fail-safe space-based assets.The inability to reach an inoperative orbital asset means it's "dead" from an investment standpoint. The ability to reach it quickly to repair it turns on the cash flow stream once again and lowers the burden of capital and fixed operating expenses. This, in turn, allows a lower cost to the ultimate beneficiary of the space-

Comes the Revolution . . . .

•

15

based operation and a greater return on investment to the space asset owner. Going back to the opening question: Does the operation ofSSTO spaceships represent an attractive business opportunity? From a fi, nancial and investment point of view, the answer is an overwhelm, ing YES! Therefore, an excellent case for profitable space transportation can be made. The financial and economic numbers will continue to be persuasive, no matter what sort of space vehicles are used, as long as they are totally reusable single,stage,to,orbit types offering turnaround times of a few days. The technologyused to do this is imma, terial. This is an important point to keep in mind. When a prospective space transportation company executive goes to New York City to talk to CitiBank, Prudential Bache, Morgan Trust, or whoever can put together the financing necessary to build spaceships or operate a spaceline, the banker doesn't care what the technology is. If Mc, Donnell Douglas, General Dynamics, or Rockwell International needs a line of credit to build spaceships, they have a track record of making similar technologies work profitably. Thus, the financier assumes the company knows what it's doing with an arcane technol, ogy he doesn't need to learn about. It's also easier if the operator has a track record. However, Fred Smith at Federal Express had no such track record when he started, just an outstanding and convinc, ing business plan, and a large inheritance. By looking at just the economic numbers, one fact should be obvious. The first people who build and operate SSTO spaceships will dominate space for the first half of the next century.

SP ACE MARKETS Determining a market size for SSTO spaceship services is uncertain since the market demand curve is largely unknown. One can logically and reasonably assume on the basis of historical evidence that a sharp drop in space transportation costs will produce a proportional increase in the demand for space transportation services. An SSTO spaceship transportation cost that is 5% of existing services indicates

16

•

Islands in the Sky

a potential annual requirement of 40 million pounds to low-Earth orbit by the year 2010. Many factors can affect this estimate for SSTO spaceship launch services. As the aircraft and airline industries learned during the 1930s, estimating errors are more likely to result from overconserva, tism when new technologies and major shifts of economics are intro, duced to a market. A look at some present-day air cargo figures can help put the estimated 40-million-pound launch forecast of 2010 into perspective. In 1988 the three major airports of the New York metropolitan area processed a combined total of more than 1.8 million tons of air cargo.3 That's 3.6 billion pounds of cargo for just one region of the world. Assuming that the air cargo market continues to grow at its steady historical rate of 2.5% per year, by the year 2010 the air cargo volume for the New York area would be about 6.2 billion pounds. The estimated 40 million pound LEO launch requirement of 2010 represents less than 1% of the projected air cargo volume of only one region of the world. Considering the historical perspective provided by air cargo, it's not fantasy to speculate that the annual space transportation services required during the first two decades of the next century would grow far in excess of the 40 million pounds of payload estimated initially. How much in excess? This is truly anyone's guess, but the numbers could be in the hundreds of millions of pounds. Obviously, this wide range of possible futures has a bearing on spaceship fleet size and the required ground infrastructure needed to support this magnitude of operation. Assumingthat each SSTO spaceship is capable of carrying 20,000 pounds to low-Earth orbit and has a flight' rate of 100 flights per year, the total fleet sizerequired to meet an estimated market demand of 40 million pounds of payload to LEO is 20 spaceships. It's unreasonable to assume that each spaceship will have a full cargo bay on every flight. For purposes of this analysis, the result is a fleet size requirement that remains at 20 spaceships. For a market requirement several times larger-such as 120 million pounds of payload to LEO-spaceship fleet size would increase to 60. Does this sound unbelievable considering the level of space transportation today? Not really. Orbital space promises to be an ideal environment for the manufacture of many high-technology products such as electronics and pharmaceuticals.4,S,6

Comes the Revolution . . . .

•

17

But some of the best products from an investment standpoint will undoubtedly turn out to be mundane or appear to be "frivolous." As the cost of access to low-Earth orbit decreases, the importance of the transportation portion of a space-based investment decision becomes less important. A look at the manifests of the air cargo carriers in and out of the New York area shows that they consist mostly of such "low-tech" items as clothes, toys, games, vegetables, fruits, and seafood. 7 To say that space will forever remain the domain of high technology is to ignore the business lessons of history. When payloads heavier or larger than those capable of being deployed by a single spaceship are considered, the cost per ton is not a smooth curve. Figure 1.1 is familiar to people in the airline business. On the vertical axis is cost per ton where in the airline business it is normally cost per passenger. The spikes are caused by reaching the maximum payload capability of a single spaceship where a customer wants to put up 11 tons. This creates a spike in the curve because an additional spaceship must be scheduled either to lift the additional ton or to divide the payload equally between two spaceships. This obviously increases the cost per ton ( Figure 1.1). But when the cost per ton of orbiting large payloads in excess of the maximum payload of a single spaceship is compared to the same costs of lifting them on the existing fleet of expendable and semi-expendable vehicles such as the Space Shuttle, Titan, Atlas Centaur, Ariane, and Delta, the results are depicted in Figure 1.2. This dramatically indicates that any reusable vehicle anywhere in the neighborhood of the financial characteristics and operating char-

REVENUE PER TON ($Millions)

1 Required Revenue Per 0.9 Flight: $4. 7 Million 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 Required Revenue Per Flight: $1.6 Million O5 15 25 35

REQUIRED PAYLOAD TO LEO (Tons) Figure 1.1. Required revenue per ton of payload.

18

•

Islands in the Sky

60 50 REVENUE 40 PER 30 TON ($Millions) 20

10

Expendable and Semi-expendable SSTO

-::::::~::.:::··•·4': ....-,.J,;

o.i--_....., ____ 5

15

....., ____ 25

,...... 35

REQUIRED PAYLOAD TO LEO (Tons) Figure 1.2. Required revenue per ton of payload for various vehicles.

acteristicsof the SSTO spaceship used in the model is going to force a major change in the market. Some people may object to this analysis on the basis that much the same sort of thing was claimed for the NASA Space T ransportation System, the space shuttle. However, the major difference of the EnterpriseInstitute analysisrevolves around the fact that the designers and builders of the STS didn't have to talk to New York bankers, but to people in Congress who are quite a different class of financiers. As detailed earlier, the secret of making money with any machine is to keep it working. That's how airlines counteract the $150million-dollarfly-awaycosts of a Boeing 747-400, for example. Because the NASA space shuttle has a multimonth turnaround time, requires a standing army of 40,000 or more people, and throws away a major part of the launch vehicle on every flight, it's impossible to fly it for less than ten times the projected cost using an SSTO spaceship. Now, more than 12 years after its first flight, the space shuttle is clearly seen to be an experimental vehicle that cannot possibly be commercially operated at a profit. However, like an airliner, an SSTO can be designed to be operated profitably.

THE HIDDEN MARKET Initially, transportation of cargos to and from orbit won't be the most profitable or largest market (see Figure 1.3).

Comes the Revolution . . . .

•

19

Figure 1.3. The hidden market. Here's the hidden market for initial SSTO Services. The World is shown in polar projection centered on a hypothetical U.S. Southwest Regional Spaceport. The numbered zones indicate how many minutes are required to reach any location m that zone from the Southwest Regional Spaceport at "X." Europe is about 25 mmuccs away, for example, while Japan requires a flight of only 30 minutes. Australia is between 40 and 50 minutes flying time for an SSTO.

Most space economists agree several years will pass before the space transportation markets adjust to the introduction of the new spaceships. While this is happening, however, the new spaceships will have a lot of work to do in another market that's usually overlooked. Any SSTO spaceship that can take a payload to orbit can also deliver passengers and cargo to any place in the world in less than an hour. A 15-hour subsonic flight to Tokyo becomes a flight of less than an hour using "Fractional Orbital Transportation"-take off, climb above the atmosphere, make a partial orbit, and come down to land at the destination. Furthermore, it's environmentally benign. Again, Robert A. Heinlein told us that fast transportation will pay because it always has. Businesses are keen to beat their competition because time is money. Very well, if operating SSTO spaceships can be a profitable business, thus justifying companies to design and build them, what then? Don't stop here or you'll miss other business opportunities!

20 •

Islandsin the Sky

THE NEW SPACEPORTS The new SSTO spaceships will no longer be chained to existing launch sites. A brief look at the historical background will show why• The original requirements for a site to launch large rockets were stated by Colonel Walter R. Dornberger in Germany in 1935. 8•9 He was an artillery officer who had been put in charge of the secret German project to develop a long-range artillery rocket with twice the range of the legendary Paris Gun. This rocket became the German V-2. Domberger wrote that a rocket launching center should be located on a seacoast because Germany didn't have open land on which these artillery rockets could crash. Since the German rocket program was secret and a lot of rockets blew up in those days, he wanted the facility to be in a remote location for security and safety. Domberger described the ultimate firing range for long range artillery. We're still wedded to Domberger's requirements because present space launch vehicles were developed from long-range ballistic missiles (ICBMs), the descendants of the V-2. Every space launch center in the world is nothing more than a super artillery range. When spaceships begin to operate like airplanes, they no longer need to be flown from artillery ranges. They will operate from true spaceports. The new spaceport requirements include: • Transportation access by rail, highway, and air. • A "noise-dear,, zone extending about 10 miles around the takeoff and landing site. • Clear airspace corridors for departures and arrivals of both vertical-takeoff-vertical-landing (SSTOs) and horizontaltakeoff-horizontal-landing (aerospacep lanes) spaceships. • Proximity to other transportation nodes-rail, highway, and air. • Proximity to existing natural gas pipelines; natural gas is a source of hydrogen that can be processed into liquid hydrogen rocket fuel. • Proximity to a high-capacity electric power grid. Note that remote locations are no longer considered necessary. But suppose these spaceships fall on our heads? People asked the

Comes the Revolution . . . .

•

21

same question about airplanes 60 years ago. The safety concern of flying spaceships over populated areas disappears with the advent of SSTO spaceships that don't drop boosters or lower stages in flight. They have engine,out capabilities and safe abort modes similar to aircraft. How often does a jet airliner lose all its engines leaving San Francisco International? The prohibition of flying rockets over populated areas arose on May 29, 1947, when a German V,2 flew off the White Sands range in New Mexico and landed in Ciudad Juarez, Mexico. The present, day flight safety system for ballistic missile and space vehicle launch sites evolved from the policy and procedures developed immediately thereafter by an artillery test officer in the U.S. Army Ordnance Corps, Major Herbert L. Karsch. 10 These require that no rocket or missile be allowed to land beyond the range boundaries. With SSTO spaceships, such flight restrictions are no longer necessary. The new spaceports will be located where they are needed, just like airports. The new spaceships and spaceports will require services other than fuel supplies and the facilities for maintenance and repair. Look at any major international airport to get some idea of the services anJ supporting businesses that will, within 20 years, be clustered around the new millennium spaceports. Such spaceports will become the industrial hubs of the twenty, first,century hydrogen economy because this is where large,volume liquid hydrogen production and handling facilities will be installed. 11

THE LEGAL ISSUE OF TORT LIABILITY The existence of a large civil SSTO spaceship fleet and extensive, growing spaceship operations also brings up some interesting legal issues. Tort liability is one of these. At this time, if a Northwest 74 7 airliner has an engine failure leaving Tokyo's Narita International Airport and crashes in the Ginza, the airline itself is responsible under the terms of international law. However, if any space vehicle has a similar accident in which property is damaged or lives lost, the incident is covered by the United Nations Treaty of Principles. Under the provisions of this early,day space treaty, if a Northwest Spaceline's spaceship has an accident, the company is not liable. The U.S. government is individu,

22 •

Islands in the Sky

ally and severallyliableunder the terms of that treaty.'2 Thus, the federal government has had to establish the Office of Commercial Space Transportation within the Department of Transportation. The OCST regulations covering safety inspections, clearances, permits, and approvals is long and costly for each individual space launch operation, regardlessof the type of launch vehicle used. These will become unworkable with frequent SSTO spaceship operations. Therefore, the United States needs to investigate getting that treaty amended to reflect the reality of the coming age of commercial SSTO spaceshipoperations. Or the president must revoke the U.S. ratification of that treaty. Ratifying a treaty requires the advice and consent of the U.S. Senate. But to revoke one requires only an Executive Order. This happened in the case of the Panama Canal Treaty. It led former Senator Barry M. Goldwater to sue the Carter administratinn in federal court and lose. Therefore, this set the precedent that the United States can back out of any treaty by a presidential order.

CERTIFICATION OF AIRWORTHINESS Certificationof spaceships is yet another issue that must be aJJressed under international law. At this time, every airplane that crosses an international border must have a certificateof airworthiness (C of A) issued by the country of aircraft registry.It must also carry a certificate of manufacture and a cargo manifest. This has been considerably muddied in the last 20 years because the NASA Space Shuttle orbiters originally didn't carry them. In 1981, an author named Lee Correy wrote a novel entitled Shuttle Down13 that created a stir at NASA Headquarters. The plot centered around an orbiter that had to make an emergency landing on Easter Island. The Chilean military governor maintained that the orbiter was an airplane that, under international law, had to have a C of A and other documentation, including passports, for the orbiter crew of astronauts and mission specialists. In the early space shuttle program, the orbiters and crewshad none of these. Now each orbiter carries all the necessary documentation except a C of A because

Comes the Revolution . . . .

•

23

Federal Aviation Regulations Part 21 has no provisions for certificating a space vehicle. The legal problem of certification remains unsolved. Again, this will change once commercial SSTO spaceships begin flying to orbit and back. Therefore, some effort needs to be addressed to drafting regulations in this regard.

DEFINITION OF AIRSPACE AND ORBITAL SPACE BOUNDARIES Yet another legal and operational issue arises over the definition and establishment of boundaries for airspace and orbital space. No accepted international definition of the interface between airspace and orbital space exists. Thus, the question arises: Where does airspace end and where does orbital space begin? Or will it make any difference? Indeed it does, especially from the standpoint of national security. At the present time, over the United States and most of the rest of the world, controlled airspace as defined in various federal and international rules and laws ends at 60,000 feet above mean sea level. That is as high as the authority of the FAA and other national aeronautical agencies goes when it comes to air traffic control. Who takes over from there? That's unknown at this time. The USAF awards astronaut wings to pilots who have flown above 50 miles, while the Federation Aeronautique Internationale with headquarters in Paris says that space begins at 100 km (62 miles). The Office of Commercial Space Transportation hasn't addressed the problem. It must be resolved in an international conference within 10 years.

SPACE TRAFFIC CONTROL In the coming era of expanded and extended space operations with SSTO spaceships, space traffic control will be mandatory. It is an absolute necessity from the standpoint of national security (Figure

1.4).

24

•

Islands in the Sky

I

I I

American STC Sector

..'--....,. .

\ 1

Pacific STC Sector/'

\

// Earth Orbit ............._,, ... .,Y -100-300 KM.

Atlantic STC Sector

,,--,rlow

1 ::-,i(-~~-,

\ -~~;,_: __-~

\ -~,'i;t,y I

, ..": ...'\ Geosynchronou~ Orbit I -36,450 KM. /

1

EARTH Asian STC Sector

'

European STC Sector

\.

,

'\ '

Figure 1.4. Possible LEO/GEO space traffic control zones.

Today, a nation launching a space vehicle is required to report the launch and the orbital elements to the United Nations after the fact. This poses a severe problem for national defense in the future when SSTO spaceshipsbegin operations within and through the atmosphere. For example, if you're on duty in the national aerospace defense command and detect somethingcoming over the horizon at Mach 20, you won't have time to identify it and determine what action to take before it's passed and gone. Does it have an announced flight plan? ls it the same vehicle that was flown on that flight plan? ls it the daily FedEx flight from Beijing that took off an hour late or a threat to someone other than UPS? Do you alert your ballistic missile defense system? Do you shoot or not? You won't have time to make a telephone call to get more information. You must know in advance. The thing leaving that

Comes the Revolution . . . .

•

25

ionization trail must be squawking the proper code on its radar transponder. If it isn't, you must have rules and procedures to imple, ment immediately. The principles of air traffic control must be extended into orbital space. Space Traffic Control out to geosynchronous orbit and proba, bly beyond will be required. And soon. Again, science fiction author Lee Correy, wrote a novel, in this magazine-in which space traffic control Manna-serialized 14 existed. Don't say that science has outpaced science fiction!

CONCLUSION Because the technology to create SSTO spaceships exists and can be shown to create a profitable business opportunity, we must assume that such spaceships will be built. The coming commercial space age has many facets that have been overshadowed by the effort to bring SSTO technology to operational status. However, a scenario exists where reusable, reliable, safe, responsive, and economical SSTO spaceships can be built but can't be operated because of political, diplomatic, or military concerns that weren't addressed in concert with technology development. It appears prudent to assume technological success and to address these other factors as quickly as possible. Acknowledgments The author expresses appreciation to Paul C. Hans and The Enter, prise Institute, Inc., for permission to utilize the results of their space transportation economic model.

Chapter 2 Robert M. Zubrin

The Hypersonic

Skyhook

Skyhol 1b. ~fany readers of science fiction and speculative astronau, tics have heard of such devices and why they can't possibly work. I think they can-hut let's begin at the beginning. Way hack, in 1960, the Soviet newspaper KomsomolskayaPravda ruhlished an interview with engineer Y. N. Artsutanov, 1 containing a description of a novel means for Earth to orbit transportation. In the Artsucanov scheme, a satellite placed in geostationary orbit would simultaneously extend cables down toward Earth and in the opposite direction, keeping its center of mass, and thus its orbit, constant. This procedure would continue until the lower cable reached the surface of the Earth, where it could be anchored and used to support elevator cabs. These cabs, in turn, could then be used to transport payloads up to the satellite where they could be released into geostationary orbit. Alternatively, if allowed to proceed out along the outward cable, the payloads could be released with greater than orbital velocity, so that at different stations along the outward cable, payloads could be released to proceed on trans,lunar, trans,Mars, trans,Jupiter and other trajectories of interest. This con, cept, while published in both Russian and English, was widely ignored and promptly forgotten, only to be rederived, and published by a group of American oceanographers 2 in 1966, after which it was 27

28 •

Islandsin the Sky

forgotton again. The concept was then independently discovered for a third time in 1975,by Jerome Pearson, 3 of the Wright Patterson Air Force Base, who published a series of papers going into much greater detail than the earlier authors, including derivations for system mass, tether taper configuration, and allowable limits to the rates that payloadcould be moved along the tether without exciting dangerous vibrational modes. Subsequently, the geostationary skyhook concept waswidelypublicized by Arthur Clarke, wht) employed such a device as a central feature of his novel The Founrains of Par-

adise. While offering the exciting prospect of a simple cahle car-toorbit system, the geostationary skyhook descrihcd hy Pearsun and the others sufferedfrom one little flaw: It was impossihle. The reason why it was impossible was this: If one places a load at the huttom of the geostationarytether, the bit of tether holding it must be thick enough to support that load. The next bit of tether must then he thick enough to support, not only the load, but the luad plus the bit of tether supporting it. Thus, as it proceeds from the ground towardgeostationaryaltitude, the tether must get th ickcr anJ thicker, and its diameter expands exponentially. Depending on the srrcngthto-weight ratio for the tether assumed, the final result would he that the cross-sectionalarea of the tether at the satellite would be 6 to 20 ordersof magnitudegreater than its area at its base, with similar fantastic ratios holding between the tether system mass and the mass of the payload it is required to lift. In answer to this problem, Pearson could only propose that in the future, ultrastrong materials, such as single-crystalgraphite fibers, might become available with orders of magnitude improvement in strength-to-weight over current materials, resulting in exponential reductions in taper diameter ratio and system mass. Until such futuristic materials became available, skyhook applications would have to be limited to lower gravity bodies, such as Earth's Moon, where while the problems faced by skyhook engineering are greatly reduced, the imperative for an alternative to rocket transportation is correspondingly less compelling. A potential means of bringing the skyhook back down to Earth was offered in 1977, by Hans Moravec,4 who proposed that the skyhook be placed in an orbit lower than geostationary, and the two tethers extending from it be made to rotate, or "roll," so that during its descent through the atmosphere, the tether tip would be moving

The Hypersonic Skyhook

•

29

backwards at an equal speed to the skyhook's orbital velocity. This would cause the tether tip to briefly possess a ground,track velocity of zero, at low altitude, thus enabling a payload on the ground or on a slow,flying aircraft to be transferred to the tether for a ride into space. By bringing the skyhook to a lower orbit, the length of the skyhook is reduced, creating the potential for a large reduction in system mass. This idea was subsequently elaborated and improved upon by Rod Hyde; and Pearson. 6 The realization of the mass reduc, tion potential in the rolling skyhook concept is restricted, however, by the fact that as the two tethers spin about the system center of mass, centrifugal force is generated, which is additive to gravity while a tether is in the downward position, and negative to it when the tether is in the upward position. This adds both static and dynamic loads to the tether system, requiring tether thickening and mass increases. The result is that futuristic materials are still required for tlw construction of a practical system. As an operational system, the nitating tether suffers in comparison to the geostationary tether by thl· hrc\'ity l)f potential access, and by the fact that while a payload can reach a stable orbit condition by climbing the tether to the skyhook center prior to release, and injection energies for interplane, tary \'ellKities are easily attained during release from the tether during its upward, forward,moving swing, useful injection orbits may be difficult to attain due to the lack of correspondence between the tether's motion and the proper orbital phasing for payload release ( i.e., the tether will generally not be moving through its upward arc at the right moment to throw the payload where it is supposed to go). In this article I present a third type of tether,skyhook system, which I call a hypersonic skyhook (Figure 2.1). As I show, the hypersonic skyhook has the potential of relieving the severe strengthof,materials demands, which have thus far prevented the implementation of a practical skyhook of either the geostationary or rotating varieties. The hypersonic skyhook accomplishes this by keeping the tip of the extended tether outside of the tangible atmosphere and allowing it to move at hypersonic velocities with respect to the ground. This allows the skyhook's center of mass to be lowered from geostationary altitude, reducing the length of the tether, and cutting its mass and taper ratio by many orders of magnitude. Unlike the rotating skyhook, however, the hypersonic skyhook does not spin about its satellite center of mass, but keeps both tethers aligned

30

•

Islands in the Sky

Figure 2.1. The hypersonic skyhook.

along a constant radial direction toward and away from the center of Earth. This causes the centrifugal force term along the tether to be constant in time and always antigravitational in nature This eliminates the dynamic stresses faced by the rotating tether and causes a further exponential reduction in tether mass, as the gravity term driving the tether mass expansion is negated more and more when the tether's orbital velocity increases. The result is that tether systems with massesonly one or two orders of magnitude greater than the payloadsto be lifted become possible using present,day or near,term materials, for hypersonic tethers moving with ground, track velocities in the Mach 10-15 range. Because the atmospheric drag caused by traveling through the atmosphere at such velocities is unacceptable, the base of the hypersonic skyhook must travel outside the atmosphere, perhaps at an altitude of about 100 km. Access to the hypersonic skyhook would have to be provided by transatmospheric vehicles capable of matching horizontal velocity with the tether and utilizingvertical rocket thrust to negate gravity

The Hypersonic Skyhook

•

31

during rendezvous. While the design of such vehicles is certainly a significant challenge, it is much less formidable than the develop, ment of transatmospheric vehicles with full orbital (Mach 25) capa, bility. The hypersonic skyhook, moreover, allows the payload to be sent from suborbital trajectories, not only to orbit, but to any destination in the solar system without the further use of propellant.

SKYHOOK EQUATIONS The relevant equations governing the performance of a hypersonic skyhook can be derived as follows: ( If you have trouble with equa, tions, don't worry about it. Just skip ahead to the bottom-line results, which are given in the following section.) The amount of mass in a given segment of the tether is given by dm

= pa dl

(1)

where p is the density and a is the cross-sectional area. The amount of additional load added to the tether due to this segment is given by: dF

= g dm

- ca,2(R1 + 1) dm

(2)

where g is the gravitational acceleration at the altitude of a given tether location, R 1 is the distance of the base of the tether from the center of Earth, 1 is the distance along the tether from the base, and w is angular velocity (in rps) of the tether with respect to the center of a fixed (nonrotating) Earth. The local gravitational acceleration, g, is given by (3) where g 1 is the gravitational constant at the base of the tether. Putting equations (1 ), (2), and (3) together and noting that the force, F, at any point along the tether equals ua, where a is the tether material yield stress, we can obtain an equation for how the tether cross-sectional area, a, changes with respect to its length, L.

32

•

Islandsin the Sky

(4) The solution of equation (4) is:

Equation (5) is the expression that yields the required taper ratio for an orbiting, nonrotating, tether skyhook of any given length, L. It can be seen that the tether taper ratio is an exponential function of the material density to system ratio, p/a, multiplied by a the length of the tether, L, multiplied by a term that decreases strongly as the angular velocity, C&>,increases. lt can also be shown from the lawsof orbital mechanics that: (6)

whereµ, is the mass of Earth times the universal gravitational con, stant. Until such time as either of those quantities should change, µ, = 398600km1/s2• Now, as the velocity of the tether with respect to the ground increases, so does w, which, as equation ( 6) shows, has a strong inverse relationship to the length of the tether from its bottom tip to its thickest (orbital center of mass) position. As equation (5) shows, eachof these effects tends to reduce the required tether taper so it may be expected that taken together, the ratio exponentially, resulting reduction in required taper ratio will be extremely strong. As we shall see in the presentation of results given below, this is indeed the case. A tether skyhook starts with a certain cross,sectional area, ao, at its base, which increases to a maximum cross,sectional area at its orbital center of mass.The ratio of the maximum area to aois called the tether's "taper ratio." The tether then continues outward beyond its center of mass, tapering off again as described by equation (5 ), or faster. ln general there will be an upper strand to the skyhook system that will extend outward beyond the orbital center of mass by a distance at least as great as the lower tether strand.

The Hypersonic Skyhook

•

33

DISCUSSION OF RESULTS The strength-to-weight ratio of tether materials is ordinarily given in terms of the tether's characteristic velodty (U), which is simply the square root of the strength (in pascals) to density (in kg/m 3) (U =~).Kevlar with a U of 1.2 km/s has been available on the commercial market for some time, while state-of-the-art highstrength kevlar-type materials with a U of about 1.6 km/s are now becoming available. Advanced materials with Us of somewhat over 2 km/s will probably be available in the near future. Figure 2.2 shows the taper ratios resulting from equation (5) for hypersonic skyhooks with Us of 1.2, 1.6, and 2.0 km/s. The case of a ground-track velocity of zero is simply the classic ArtsutanovPearson geostationary skyhook. It can be seen that, postulating present-day or near-future materials, such devices require taper ratios that are astronomical, making their construction infeasible. The merit of allowing the skyhook to move at hypersonic velocities can, however, also be seen. For example, assuming a U of 1.6 km/s, the geostationary skyhook requires a ludicrous taper ratio of 108, while one whose base is allowed to move at 4 km/s (Mach 12) only requires a quite reasonable taper ratio of 102. Figure 2.3 shows the mass of a hypothetical skyhook system capable of lifting loads of 4 tonnes. The skyhook mass was estimated by integrating equation (5) from the bottom of the tether to its orbital center of mass (to calculate the mass of the lower tether

.2

•...

IC

l

• ... .¥

0 0

I:

:ii,.

.¥

en

10 1 s 10 1 4 10 13 101 2 10 11 10 1 0 10 • 10• 10 7 10• 10 S 10" 103 102 10 1 100

--D

2 Skyhook

Taper

(U■1.2)

Taper(U■1.I) Taper

4 Veloclty

(U■2.0)

I (km/a)

Figure 2.2. Taper ratio for hypersonic skyhooks.

•

34

•

Islands in the Sky

ii

•

C C

!

•• •

:I ~

0 0

,..

.c

~

fl)

--

101S 10 14 1013 1012 10 11 1010 10 I 10• 10 7 10• 10' 104 103 102 10 1 100

0

MaM (U■1.2) MaM(U■ 1.I) MaM(U■2.0)

4 2 Skyhook Veloclty

8

I (km/a)

Figure2.3. Mass of hypersonic skyhook tether lifting 4-mnne payloads.

directly),doubling this account for the additional mass of the strand of the tether beyond the orbital center of mass ( this is an approximation-there is a certain amount of design freedom in the configuration of the tether upper strand), and multiplying the result by 1.25 to allow for margin. Once again, the drastic reduction in mass of the hypersonic skyhook compared to the geostationary variety is readily seen. Table 2.1 gives the lower tether length ( 100 km less than the skyhookcentral platform orbital altitude) and the total tether mass (both upper and lower strands) for hypersonic tether skyhooks capable of lifting 1-tonne loads, again with a safety margin of 1.25. It can be seen that while a geostationary skyhook with a U Table 2.1. Hypersonic skyhook performance for tonne payloads Sk:yhook TetheTMass (lonnes) Veloci1:y Lower l..mgih(km) (km/s)

0.0 1.0 2.0 3.0 4.0

5.0 6.0 7.0

35727 13275 7501 4667 2936 1751 880 208

UpperEndCJ

u = 1.2 3.68X 1016 3.23 X 1011 1.22E 8 325,000 3,575 122 31 0.33

U = 1.6 1.40X 1010 1.26X 107 117,000 3,425 224 26 3.8 0.19

U

= 2.0

1.34 X 107 102,250 4200 375 55 11 2.1 0.12

(km 1/s1)

21.9 32.1 30.0 21.4 8.1 -8.8 -29.1 -52.4

The Hypersonic Skyhook

•

35

of 2.0 km/s requires a total (upper and lower sections combined) tether mass of more than 13 million tonnes, a hypersonic skyhook with a ground-track velocity of 5 km/s only requires 11 tonnes of tether mass. Also shown in Table 2.1 is the energy measured in units of "C/' that the payload will have if released from the upper end of the tether, assuming that the tether is constructed symmetrically, with the upper strand identical to the lower strand. (The C 3 of an orbiting spacecraft is the difference between the square of its velocity and the square of the planet's escape velocity at its altitude.) Since a C 3 of 8 km 2/s 2 is about what is ~equired for a minimum energy trajectory to Mars or Venus, after which a planetary gravity assist can be obtained, it can be seen that a hypersonic skyhook with a base velocity of 4 km/s or less can be used to transfer payloads from a suborbital condition to a trajectory to any planetary destination in the solar system. While it cannot achieve Earth escape, a payload released from the 5 km/s skyhook has more than sufficient energy to achieve an elliptical orbit whose apogee is at geostationary altitude ( i.e., a GTO orbit).

HYPERSONIC SKYHOOK SYSTEM let's now take a look of what a sample early skyhook system might be like. Based on the results given in Table 2.1, we'll assume a 1.5-tonne payload delivery design with a ground-track velocity of 5 km/s and U = 2.0 km/s. The required tether thus has a mass of 16.5 tonnes. A power source will also be required to lift the payloads up the tether by a cable car mechanism; 30 kW should do it. This can be provided by either photovoltaic panels or a nuclear reactor. Assuming near-term space technology, this will have a mass of 2 tonnes for the power source and its power conditioning system. While moving up the lower strand of the tether, the 1.5-tonne payload has an initial weight of about 7800 N (the weight is reduced by centrifugal force), decreasing to zero at the orbital center of mass. If we assume an average weight of 4000 N, the average speed that the payload can be moved is (30,000 W/4000 N) = 7.5 m/s. Since the lower strand of the tether is 1751 km long, it will thus take about [1,751,000/(7.5*3600)] = 65 hr to move the payload from the

36

•

Islands in the Sky

tether lower tip to orbital altitude at the skyhook center of mass. Raising the payload beyond the orbital center so that it can be released at some point along the upper strand of the tether requires no power at all, as beyond the orbital center of mass, the outward centrifugal force term along the tether is greater than the inward pulling force of gravity. Effectively, a payload being transferred out along the tether upper strand is being lowered away from the Earth. While the bottom of the tether has a ground-track velocity of 5 km/s, it has a velocity of 5.471 km/s with respect to a fixed (nonrotating} Earth (assuming a tether bottom altitude of 100 km.). The tether center of mass is moving at a velocity of 6.96 km/s with respect to a nonrotating Earth. Moving a 1.5-ronne payload from the tether base to its orbital center of mass thus increases the payload's angular momentum about Earth by (6.96*8229) (5.471*6478) = 21,883 tonne-km 2/s. Since the angular momentum of the total skyhook-payload system about Earth must be conserved, thrust will be required or the skyhook will lose altitude. If we adopt the approximation that the total mass of the skyhook (about 20 tonnes, accounting for various miscellaneous subsystems and micrometeorite protection) is concentrated at its orbital center of mass, we find that a total /lV of [21,883/(20*8229)) = 0.134 km/swill be required for the tether to maintain altitude. If this is provided by ion engines with a specific impulse of 5000 s, then 54 kg of propellant will have to be expended to enable the skyhook to maintain altitude after the lifting of each LS-tonne payload. Alternatively, part of the tether could contain an aluminum wire whose current, interacting with Earth's magnetic field, could be used to produce thrust without the expenditure of any propellant. Based on existing studies,7 a 20-km-long electromagnetic (10 km above the orbital center of mass and 10 km below) tether operating in an equatorial orbit at a distance from Earth's center of 6778 km and consuming 30 kWe would have a mass of about 800 kg and be able to generate a thrust of 3.75 N. Our skyhook operates in an equatorial orbit at a distance from the center of Earth of 8229 km, and since Earth's magnetic field falls off as the distance from its center cubed, this implies that such a design would exert a thrust of 2.1 N if used to propel the skyhook. This amount of thrust would be sufficient to provide the required compensatory fl V for the skyhook in about 15 days. If more frequent use of the skyhook were desired, part or all

The Hypersonic Skyhook

•

37

of the d V could be provided by lifting a high-thrust propellant to the orbital center of mass as part of the payload. For example, if hydrogen/oxygen propellant with a specific impulse of 450 s were used to provide half of the required 134 m/s d V, then 306 kg of the 1500-kg payload would have to be propellant, but the tether could provide the rest of the reboost electromagnetically in a period of 7.5 days, thus allowing it to be used about 40 times per year. The skyhook system could be made safe from severance by orbital debris by using a multistrand configuration such as that recently proposed by Robert Forward. 8 Payloads would be delivered to the skyhook by a transatmospheric vehicle, perhaps similar to the National Aerospace Plane (NASP) or Single-Stage-to-Orbit (SSTO) vehicles currently being developed. Vehicle performance requirements would be far more modest, however, since only suborbital (Mach 15) flight would be needed, instead of the Mach 25 speeds entailed for both flight to orbit and orbital reentry. The transatmospheric vehicle would match speeds with the skyhook bottom ( which is hanging at an altitude outside of the tangible atmosphere) and use vertical thrusters to negate gravity during the period of rendezvous. During rendezvous the transatmospheric vehicle would hover below the tether, open its cargo bay and allow its payload to be hooked by a cable car mechanism, which rides upon the tether. After the payload is hooked, the transatmospheric vehicle would drop away, close its cargo bay, and return to Earth. Assuming that 30 s of thrust-negated gravity are required for the hooking operation, and hydrogen/oxygen thrusters with a specific impulse of 450 s are used, then an amount of propellant whose mass is about 3.3% of the transatmospheric vehicle will have to be expended during rendezvous. The amount of fuel used is only 3.3%, not 6.6%, because a suborbital vehicle flying outside the atmosphere at a ground-track velocity of 5 km/s only experiences about half of Earth's gravity. (This suggests another interesting use for hypersonic skyhooks: Use them to suspend nonrotating, partial-gravity space stations. Such stations could be used to train crews or test equipment under simulated lunar or Martian surface gravity conditions.) Using its own electrodynamic thrust capability, the skyhook could deliver itself from LEO to its design altitude of 1851 km in about 86 days, without requiring any propellant. The system's total

38

•

Islands in the Sky

initial mass in LEO would thus be the same as its operational mass, that is, about 20 tonnes, and either a Titan IV or a space shuttle could lift it to orbit in a single launch. Each 1.5-tonne payload lifted to the base of the skyhook would be delivered not merelyto orbit, but, by running out along the upper tether of the skyhooksystem, could be placed on a GTO trajectory. The deliveryof 1.5 tonnes of OTO represents an equivalent capability to that represented by a Delta/Pam launch vehicle and upper stage combination, which has a ground liftoff mass of 220 tonnes, and a cost of $60 million. A single-stage suborbital rocket using HJ0 2 propellant with a trajectory averaged lsp of 420 s and a stage dry mass fraction of 20% that could deliver the LS-tonne payload to the skyhookbase (6 km/s aV assumed), hover for 30 s, and land, would have a ground liftoff mass of 25 tonnes. Thus, the skyhookaided system would only have a liftoff mass of 11% of that of the Delta and could be fully reusable as well. Savings per mission would probablyexceed $55 million, with financial breakeven against the $130 million required for the initial skyhook Titan launch exceeded on the third suborbital launch.

STORMING HEAVEN The near-term hypersonic skyhook described previously can thus reduce the cost of launching medium-sized satellites by about an order of magnitude,but in the longer term, it represents something far more important. The skyhookis a beachhead for the storming of heaven. Let's see how this could work. The skyhook described above has a mass of 20 tonnes and can assist in the lifting of 1.5 tonnes to orbit per suborbital launch. This means that in 14 launches of a reusable suborbitalvehicle, the mass of the skyhook can be doubled. If the parts lifted to orbit are simply a parallel skyhook system, the two can be made to work in tandem, and we now have a skyhook capable of lifting 3 tonnes to orbit per suborbital launch. The skyhook's capability has been doubled, and the total ground liftoff weight of the 14 reusable suborbital launches required to double it, about 350 tonnes, is less than half that of the 850-tonne expendable Titan IV (or 2000-tonne partially reusable space shuttle) required to launch the initial system.Using 14 launches of a larger suborbital

The Hypersonic Skyhook

•

39

vehicle, with a ground liftoff weight of 50 tonnes, the skyhook can now be increased to a system with the capability of lifting 6 tonnes per launch, with similar savings. We could go on like this, simply using the leverage offered by the 5-km/s skyhook to double its capacity with every 14 launches, or we could use some of the mass lifted to extend the skyhook so that instead of requiring 5 km/s groundtrack velocity for the suborbital launcher, only 4 km/swill be needed. If the AV required to achieve orbit is dropped in this way, the payload delivery capability of the suborbital system will be doubled. This will cut in half the cost of delivery of all further mass to orbit. This is only the beginning. As the skyhook continues to operate in support of its own expansion, the mass of payloads deliverable to orbit can be made to increase exponentially, while the costs of delivery drop in inverse proportion. As the skyhooks are lengthened to allow for flight to orbit at lower and lower velocities, the delivery of large payloads to orbit using hypersonic airbreathingsystems will become increasingly practical. Within a comparatively short time, transporting people and payloads from Earth's surface to orbit could become as cheap as long distance jet travel is today. The golden door to the high frontier will then be wide open-and not just for governments with billions to spend, but for anyone willing to fly. For the teeming masses of humanity, the Space Age will have truly begun.

Part Two

Stepping Into the Solar Systetn

Chapter 3 Robert M. Zubrin ancl David A. Baker

Mars Direct A Proposal for the Rapid Exploration and Colonization of the Red Planet Can the Unired States send humans to Mars during the present decade? Ahsolutcly. We have developed a set of vehicle designs and a mission architecture that can make this possible. We call the plan Mars Direct. The Mars Direct approach does not require any in,orbit assembly or refueling, or any support from the Space Station or other orbital infrastructure. However, despite this, the Mars Direct plan is not merely a "flags and footprints" one,shot expedition, but puts into place immediately an economical method of Earth-Mars transporta, tion, real surface exploratory mobility, and significant base capabili, ties chat can evolve into mostly self,suffi.cient Mars settlement. In this article we present both the initial and evolutionary phases of the Mars Direct plan. In the initial phase, only chemical propulsion is used, sending four persons on 3,year round,trip Mars exploratory missions. Two heavy,lift booster launches are required to support each mission. In the second phase of Mars Direct, nuclear thermal propulsion is used to cut crew transit times in half, increase cargo delivery capacity, and create the potential for true global mobility through the use ofCOz,propelled ballistic hopping vehicles (NIMFs). 43

44 •

Islands in the Sky

During the second phase of Mars Direct, as the exploration of the planet accelerates, a new strategy is also introduced, leading to the rapid establishment of a 48-person permanent Mars base. This base, in tum, acts as a construction team to prepare the economic infrastructure required to support a wave of colonists to follow. We shall show that the potential exists to colonize Mars during the twenty-first century at a rate comparable to that of European settlement of North America during the seventeenth, and that the colony so established can not only survive, but prevail, ultimately transformingMars into a world habitable by humans and much of our kin of the terrestrial biosphere.

MARSDIRECT:PHASE I The implementation of the initial phase of the Mars Direct 1•2 plan is shown schematically in Figure 3.1. In 1997 a single shuttle derived "Ares" heavy-lift launch vehicle (Figure 3.2) lifts off from Cape Canaveral and uses its upper stage to fire a 40-metric-ton unmanned payloadoff on a minimum energy trajectory to Mars, where it aerobrakesand lands. This unmanned payload consists of the following:

Figure 3. 1. The Mars Direct missionsequence. Every 2 years two boosters are launched. One sends an unmanned Earth return vehicle (ERV) to fuel itself at a new site, the other sends a manned hab to rendezvouswith an ERV at a previously prepared site.

Mars Direct

•

45

Ares Launch Vehicle Definition .e,v1o■ d Cap ■ blllllr• (All w .. gnis ,n 1onnes) Trans-Mars (C,. IS km hec:1) 47.2 Tr1n1-Lun ■ r (S aay uansler) 59.1 LEO (l60by l60N,..._28.5degrHs) 121.2 LEO/NUS (160 by 160 Nml. 28.S deg) 75.0 tsJA!!l(m)

12.3

grpu

2.114.1

M ■ ,a 1w,11,ou, Payload)

lllAt:l!

2 Advanced

Solld Rockel Boo■let ■

1,214.S

S1Ur.;1

E11ern■ I Tank (lnc1uc:1,ngResiduals) SSME Engine Pod (4 SSME·s) UHble Propell ■ nl In ET To1el SSME Thrutl (kN. 104%) Speclllc Impulse (sec) S11glng Rel ■ llve Veloclly (kmlS) (LEO 10 Mars Range)

35.15 28.6 723.5 8,7015 453 4.2 lo 5.S

i1G.cl

(lgnlled Sub-Orb1lal) Usable Propellant tnett Mats Single Engine Thrutl(kN) Specthc lmpulte (sec)

payio ■ d

fll!IDA

IALS 0■ 11gn)

151.1 13.2 1,113

465 20.4

Figure 3.2. The Arcs launch vehicle can send 47 tonnes on direct trans-Mars injection, 59 tonnes to trans-Lunar injection, or 121 tonnes to LEO.

( I) an unfueled two-stage ascent and Earth return vehicle (ERV) employing methane/oxygen engines and including a life-support system and enough whole food for four people for 9 months, plus some dehydrated emergency rations; (2) 6 tons of liquid hydrogen; (3) 100-kilowatt-electric (kWe) nuclear reactor mounted within a small methane/oxygen internal combustion driven unpressurized utility truck; ( 4) a small set of compressors and automated chemical processing unit; and (5) two small robotic rovers. As soon as the payload is landed, the 100-kWe reactor is driven a few hundred yards away from the landing site and lowered off the truck into either a natural depression in the terrain or one created by the robots with the aid of an explosive charge. Its radiators are then deployed and cable run back to the lander. Then, the reactor, which has not yet been used, is started up to provide 100 kW of electric power to the site facilities. The compressors are then run to acquire carbon dioxide out of the Martian atmosphere ( which is 95% CO 2). With the help of a catalyst, this CO 2 can be made to react with the 6 tons of hydrogen cargo, transforming it in a few days into 39 tons of methane and water. This being accomplished, we no longer have to deal with the problem of storing cryogenic

46

•

Islands in the Sky

liquid hydrogen on the surface of Mars. Next, the chemical plant goes to work, electrolyzingthe water into hydrogen and oxygen. The oxygen is stored as liquid oxygen, and the hydrogen is reacted with more CO 2 to create more methane and water, and so forth. Additional oxygen is produced by directly decomposing atmospheric CO2 into oxygen and carbon monoxide, storing the oxygen and dumping the CO. In the course of 10 months, about 108 tons of methane/ oxygen propellant is produced. While this may appear somewhat involved, the chemical processesemployed are nineteenth-century technology. The 100-kWe reactor isn't, but practical nuclear reactors have been operated since 1954, and the SP-100 in particular is currently scheduled to be groundtested in 1995,so that with the help of an accelerated program it can certainly be made ready in time for this mission. It took 8 months for the unmanned payload to fly to Mars, plus 10 months to manufacture the propellant, for a total of 18 months. Since Mars launch windows occur at 26-month intervals, there is thus plenty of time for flight controllers to assure themselves that an adequate supply of methane/oxygen bipropellant has been produced to support crew surface operations and Earth return. If it has, in 1999 two more Ares boosters will lift from the cape within a few weeksof each other. One of them is an unmanned payload identical to the one launched in 1997. The other is a manned spacecraft looking somewhat like a checker "king," 8.4 m in diameter and 4.9 m tall. Its two decks contain some 110.2 m 2 of floor space, allowingit to accommodate a crew of four. With a mass of 40 metric tons (including parachute, landing propellant, provisions, and a pressurized methane/oxygen internal combustion-driven ground car) it is light enough that the booster upper stage can project it directly onto a minimum energy (C 3 = 15 km2/s2) transfer orbit to Mars without any Earth-orbit refueling or assembly. Once on its way to Mars, the manned habitat pulls away from the expended booster upper stage that launched it, but they are still connected by a tether 1500 m long. With the help of this tether, the empty upper stage can be used as a counterweight, and the assembly is spun up at l rpm to provide a level of artificial gravity equal to the 3/8 gee found on the surface of Mars. When the manned craft arrives at Mars, the tether and upper stage are discarded, and the ship lands in the immediate vicinity of the now fully fueled

Mars Direct

•

47

ascent vehicle that has been waiting for it since 1997. The landing is safe because the robotic rovers sent out in the advance landing have identified and given extensive characterization of the best landing site in the vicinity, and photos of the site will have been used for astronaut training for the final touchdown. The ERV has a radio beacon much like an ILS transmitter at an airport, giving the crew exact position and velocity data during their approach and terminal landing. In 1976, the United States sent two Viking probes to Mars, and landed them within 30 km of their designated target sites. With the help of the landing beacon, superior technology, advance meteorological data from the ground site, and the on-the-spot decision making capability of a human pilot, we can vastly exceed the degree of landing precision demonstrated by Viking and land within a few meters of our desired touchdown point. But even if we missed by a considerable distance, the mission plan has built into it three layered fallback options, a defense in depth to assure the safe return of the crew. First, the manned spacecraft carries with it a methane/oxygen driven pressurized rover with a one-way range of 1000 km, so if worst came to worst, the crew could still drive over to their return vehicle. Second, if by some inconceivable mischance the crew misses its landing site by a distance greater than 1000 km, they can still direct the second unmanned payload ( which has been following them out a few days behind) to land near them. It contains a propellant factory of its own and can thus act as an emergency backup. Finally, if all else fails, the crew has with them in their habitat enough supplies to last them until a relief expedition can be sent out 2 years later. However, assuming that the manned landing has been carried out correctly at the prepared site, and the flight readiness of the 1997 ERV is verified, the 1999 unmanned lander will be directed to a second landing site 800 km away from the first. There it will begin manufacturing propellant for the second manned mission, which will be sent out in 2001. Thus, each manned Mars mission requires just two Ares booster launches; one to deliver a ride home, and the other to create a new outpost or add to an existing base on Mars. This is a much more economical mission than a conventional plan in which all the propellant is brought from Earth, which typically requires four to seven

48

•

Islandsin the Sky

Satum-V class booster launches each. The Mars Direct mission plan is superior to the conventional plan in another respect: Mars Direct brings all of the crew and their hardware to the surface where they can do their job of exploring Mars. The conventional plan requires leaving a mother ship in orbit around Mars, whose crew will accomplish little except to soak up cosmic rays. The crew on the surface is protected by Mars'satmosphere from most of the solar flare hazard, and with the help of some sandbags placed on top of their landed habitats, can be protected from cosmic rays as well. The vulnerability of the crew of the orbiting mother ship tends to create an incentive to limit the stay time of a conventional mission at Mars. This leads to very inefficient missions. After all, if it takes 1.5 years of roundtrip flight time to travel to Mars, a stay at the destination of only 30 daysseems rather unrewarding. A not too rough analogy to such a mission would be planning Christmas vacation in Hawaii but arrangingthe itinerary to include 9 days of transferring around airports going out and back, and half a day at the beach. Worse yet, in their rush to get back from Mars, the conventional mission planners are forced to take disadvantageous high-energy orbits, which drive up the propulsion requirements and thus the mass of the mission, as well as a swing-by of the planet Venus where the Sun's radiation is twice that at Earth. In the plan we offer, the crew will spend 500 days on the surface of Mars, coming back via the most efficient,"minimum energy" orbit possible. During their 500-day stay on the surface of Mars, the crew will be able to accomplish a great deal of exploration. Using 12 of the 108 metric tons of methane/oxygen propellant to power their ground car, they will be able to travel over 20,000 land km at speeds of over 30 km/hr, ranging out from their base 500 km in any direction. If a condenser is added to capture for later recycling the water vapor in the ground car engine exhaust, the 20,000 land km available to the ground car can be increasedseveral times over. Once the second lander's propellant production operation is well underway, they can even drive over to use it as a second base for forays. Thus, about 1.5 million km2 of territory will be available for exploration for the first mission crew alone. With a crew of four, a large landed habitat/laboratory, and a substantial power source, a large variety of scientific investigations can be accomplished. In addition to searching for past or present life and clues to the planet's geologic history,

Mars Direct

•

49

one key item on the exploratory party's agenda will be to locate pockets of readily exploitable water ice. Once native water is available, it will no longer be necessary to ship hydrogen from Earth, and future missions and settlements can be made independent of Earth for their transportation and life support consumables. But even on this first mission, an inflatable greenhouse can be set up and extended experiments undertaken in growing food crops. If successful, the greenhouse can even be left in operation after the crew departs, allowing research to continue telerobotically from Earth, and perhaps providing future crews with both food and earthly fragrances. An artist's conception of what the Mars Direct surface base would look like is depicted in Figure 3.3. At the conclusion of the 500 days on the surface, the crew will climb into the methane/oxygen ascent vehicle and rocket back directly to Earth, where they will aerobrake into orbit and either rendezvous with the space station or be picked

Figure 3.3. Mars Direct surface base. Shown is the two-deck disc-shaped hab module,

the two-stage conical methane/oxygen-driven ERV, an inflatable greenhouse, and a pressurized ground rover. A 100-kWe nuclear reactor positioned in the crater in the background has long since completed its job of driving the production of the ERV's propellent supply and now provides a copious source of power to the base, with backup power available from solar panels. (Painting by Robert Murray.)

50

•

Islands in the Sky

up by a shuttle. Quarters aboard the ascent vehicle will be somewhat cramped, but no more so than in the space shuttle. The return trip will be carried out under zero-gravity conditions, but it will only last about 6 months, and Mir cosmonauts have proven that zerogravity exposureof such length can be tolerated by humans without excessive bodily harm. Not too long after the mission 1 crew has departed Mars, the mission 2 crew will arrive and land their habitat near the unmanned ascent vehicle that had been sent out following the mission 1 crew in 1999. Accompanying them will be a third unmanned ascent vehicle/fuel factory payload which will be landed at a new site 800 km further along, to be used for return by the mission 3 crew, which will depart Earth in 2003. Thus, every 2 years a new base will be establishedand its vicinity explored, and before long a string of smallbaseswill dot the map of Mars, separated by distances within the capability of available ground transportation. With just two boostersbeing launched every 2 years, the total launch requirement needed to sustain this program of exploration averages to only one launch per year! The Mars Direct vehicle systems can also be used to accomplish lunar missions in the following way.,First, an Ares booster launch is used to throw a 59-metric-ton paylbad consisting of a standard hab module plus a cryogenic lunar orbital capture and lunar descent (LOC/LD) stage onto trans-lunar injection. The LOC/LD stage is then used to land the hab on the Moon. After one or more such habs have been thus enplaced at a given site, the crew is flown out to the Moon within a Mars Direct ERV. The ERV in this case has its (Mars ascent) first stage deleted, but its second stage is fueled with methane/oxygenbipropellant, and this provides sufficient thrust and fl V for an Earth return direct from the lunar surface. Landing on the Moon at the prepared site is accomplished with the aid of the same cryogenic LOC/LD module used to land habs and cargo. After landing, the crew exits the ERV and enters the prelanded hab(s) and proceedsto operate on the lunar surface for an extended period, after which they reenter the ERV and execute a direct return to low Earth orbit (LEO). Prior to lunar liquid oxygen {LOX) becoming available, the mass of the fully fueled Mars Direct ERV exceeds the lunar delivery capability of a single Ares launch by S tonnes. This problem could

Mars Direct

•

51

be resolved by scaling down the lunar ERV by 20% compared to the Mars version. However, if hardware commonality with the Mars ERV is desired, a simple solution would be to preland a cargo flight of liquid oxygen at the chosen site. Since such a cargo flight could land about 21 tonnes of LOX, one such flight could support four manned missions to that destination ( the ERVs would fly out fully fueled with merhane and 2/3 fueled with LOX), plus any number of cargo flights. At the conclusion of these missions, lunar LOX production could be in place and lunar LOX available at the site, eliminaring the need for any further LOX delivery flights. This lunar mission architecture has many advantages. First of all, no lunar orbit rendezvous (LOR) is required. This is almost essential for a lunar base mission (as opposed to Apollo short-stay missions), as it would be unthinkable to have astronauts exposed to zero-gravity and cosmic radiation in a lunar orbiting craft while awaiting the return of a 6- to 12-month duration lunar excursion crew, and on the other hand landing the entire crew would leave the vital Earth-return mother ship in an unstable lunar orbit for an extended period with no one minding the store. If its condition were to deteriorate, only limited corrective action could be taken and the crew might be stranded; if a failure on board were to remain undetected until LOR, the returning crew would find itself in a very difficult situation. Second, the use of methane/oxygen for direct return from the Moon has advantages compared to either the hydrogen/oxygen or conventional storable bipropellant (NTO/MMH) alternatives. Compared to hydrogen, methane is almost indefinitely storable on the lunar surface, thus facilitating extended stays. Methane/oxygen has a higher specific impulse than NTO/MMH (373 s compared to 343 s); however, once lunar oxygen is introduced, this advantage becomes greatly multiplied and the effective lsp of the methane/oxygen system exceeds 1700 s. What this means is that using NTO/MMH, 25 tons must be landed on the Moon to return a 10-ton ERV to LEO, while using methane and lunar oxygen, only 13 tons need be landed. Thus, 12 additional tons of cargo can be flown to the lunar surface with every manned flight, which means that if one crew and one cargo flight are flown each year, the total cargo delivered has been increased 50%. Finally, the Lunar Direct architecture has the key advantage of being totally coherent with the Mars Direct architecture, using the same vehicles. Thus, experi-

52

•

lslands in the Sky

ence with Mars systems can be obtained in near,Earth space on lunar missions, and overall Space Exploration Initiative (SEI) pro, gram costs can be greatly reduced through reduction in the number of elements of the total space transportation architecture. The complete set of vehicles required for this combined Lunar/ Mars Direct architecture is shown in Figure 3.4. While no detailed

ETOVehicle

Lunar Vehicles

Return Vehicle

Hab

MarsVehicles

Return Vehicle

Hab

Figure 3.4. The complete space transportation system required to accomplish both the lunar and Mars objectives of SEI using the basic Mars Direct system architecture. The high degree of vehicle commonality can help keep SEI costs to a minimum.

Mars Direct

•

53

costing of this architecture has been done, it is difficult to imagine how a SEI based on such a limited number of hardware elements could cost anything resembling the $400 billion price tag frequently cited for more conventional architectures, including as they do orbiting space stations, cryogenic depots, and a raft of specialized space transfer and lunar and Mars excursion vehicles.

MARS DIRECT: PHASE II NTR for Increased Payload Delivery At some point after the commencement of this program, a new technology, nuclear thermal rockets (NTR), 3 will come into use that will allow us to greatly increase the payload transferable to Mars with each launch. It is the advent of such technology, combined with a reorientation of mission strategy from simple exploration to exploration plus base building, that marks the transition of the Mars Direct plan from Phase I to Phase II. In keeping with the overall approach of the Mars Direct architecture, namely the elimination of the need for orbital infrastructure, NTR technology is incorporated into the plan simply as a third stage to the Arcs launch vehicle. Such a configuration is shown in Figure 3.5. The NTR stage has a specific impulse of 900 s, a power of 900 Megawatts-thermal (MWth), and a thrust of 45,000 lb. With the addition of this stage the Ares can throw 70 metric tons onto a minimum energy trans-Mars injection (TMI) trajectory (C3 = 15 km 2/s 2), an increase of about 50% over the 47-metric-ton TMI all-chemical Ares baseline. The performance of the NTR augmented Ares is limited by the fact that the NTR must be fired from LEO, and the baseline 2stage Ares vehicle is not optimized for LEO delivery, as with 250,000-lb thrust its second stage is underpowered. (The 250-klb thrust is optimized for direct trans-Mars injection). If the thrust of the second stage is increased, the LEO delivery capability can be increased, and the TMI throw also increased accordingly. Such results are shown in Table 3.1. In this calculation, NTR thrust to weight ratio of 5 and tank fractions of 0.1 are assumed. Thus, if we augment the Ares upper stage to maximize its capability as a LEO delivery system, about 83

54 •

Islands in the Sky

t--1.1111

I/

J.JI• TR Pro ell1n1

,..,.. ,.....

t--+--~-i U•

Figure3.5. Ares with NTR third stage.

tonnes can be thrown to Mars with each launch (excluding the mass of the NTR stage, which is expended). This is about 77% higher than the Ares baseline, a TMI mass increase of 35 tonnes, which translates into an extra 20 tonnes of cargo delivered to the Martian surface with each launch. If we stick with our early plan of two launches per mission, this will allow us to increase our crew complement of each flight to 12. Alternatively, if the sizeof the missions is kept the same, using NTR will allow us to launch each mission with a single booster, instead of a split between two. As a third alternative, crew sizes can be kept at four for each two-launch mission, but large amounts of cargo landed with them, Table3.1. NTR augmented Ares TMI throw capability

Stage2 Thrust(klb)

LEO Payload( tonnes)

TMl Payload( tonnes)

250 500 1000

121

70 76 83

130 144

Mars Direct

•

55

allowing for the rapid buildup of a sizable base. Finally, if fast interplanetary transits are deemed important, the NTR stage can be used to throw a four-man Mars Direct hab onto a fast (80- to 120-day transit, C 3 = 50 km 2/s2) trajectory to Mars, while using its larger minimum energy TMI throw to deliver an ERV augmented with a third stage to the Martian surface. With the help of this third stage, the methane/oxygen-driven ERV is able to execute an equally fast Earth-return trip.

NIMF VEHICLES FOR GLOBAL MOBILITY NTRs can also be designed to use Martian CO 2 as their propellant. Since this can be acquired at low-energy cost through direct compression out of the atmosphere, rocket vehicles so equipped will give Mars explorers complete global mobility, allowing them to hop around the planet in a craft that can refuel itself each time it lands. Such a vehicle concept, known as a NIMF4·5•6 (Nuclear rocket using Indigenous Martian Fuel) is illustrated in Figures 3.6 and 3.7. The NIMF vehicle shown in Figure 3.7 has a 2513-MWth engine that generates 436 klb of thrust at a specific impulse of 264 s by

}pf~ Ascend to Orbit or Direct Interplanetary Transfer Landing~! Vehicle ~

Nuclear Rocketplane (future option)

Ballistic Hopper Refuels with Martian CO2

~ Power for Mars Ba~ (future option) se

~ l__J

~

Figure 3.6. The NIMF concept in a variety of operational modes.

56 r'{"-

• - .. • ..

Islands in the Sky .,,-,,

-

.

T..,.•,.

-

••..

_.,. .. ,.

..--

--·-

---

Figure3.7. Ballistic NIMF vehicle on the Martian surface. From top to bottom we have the control declc, habitation declc, compressors, main propellant tank, and the NTR engine with coaxial tanlc wrapped around it for extra shielding on the surface. (Painting by Robert Murray.)

heating CO2 to a temperature of 2800 K and exhausting it out a nozzlewith an expansion ratio of 100. (The high molecular weight of CO2,while very detrimental to specific impulse, allows for a much higher thrust to be generated by a NIMF engine operating at a given power level than a conventional hydrogen-fed NTR of the same power). The vehicle, which has a dry mass of 45 tonnes, can contain a total of 346 tonnes of CO2 propellant, giving it the capability of attaining a 250-km by 33,000-km (250, by 1-sol) elliptical orbit about Mars, or more important, enabling it to hop from one point on the surfaceof Mars to any other point in a single hop. In addition, the NIMF can carry significant amounts of cargo over continental, scale distances, as shown in Table 3.2. Because CO2 becomes an oxidizing medium when heated to elevated temperatures, conventional NERVA 7 (Nuclear Engine for

Mars Direct

•

57

Table 3.2. Cargo delivery capability of the ballistic NIMF

Cargo ( tonnes)

Mass Ratio

ll.V (km/s)

Range (km)

10

7.03

Orbital

20 30

6.04 5.23

40 50

4.61 4.13 3.73

5.047 4.65 4.28 3.96 3.67

3000 2280 1800 1450 1220

60 70 80 90 100

3.41 3.13 2.90 2.70

3.41 3.17 2.96 2.75 2.57

8500 6000 3920

1000

Rocket Vehicle Applications, developed during the 1960s) type carbide fuel elements cannot be used in a NIMF engine. Instead either oxide or oxide-coated fuel pellets would have to be used. Uranium-thorium oxide has a melting point of about 3300 K, and such pellets coated with a layer of either zirconium or thorium oxide to retain fission products could well enable operation at 2800 K. Alternatively, preliminary data 8 indicates that "traditional" NERVA uranium carbide fuel elements coated with graphite can have their graphite coated with a further layer of thorium oxide, and that such thorium oxide outer coatings are resistant to both CO 2 and solidsolid reactions with the graphite at temperatures up to 3000 K. Because the actual propellant temperature attainable in NIMF vehicles is unknown at this time, it is useful to parameterize NIMF performance as a function of propellant temperature. Such results are shown in Figure 3 .8. Since the NIMF vehicle shown in Figure 3.7 has a maximum mass ratio of about 8.7, we can see in Figure 3.8 that a propellant temperature of about 2600 K is required if the 250-km by 1-sol orbit is to be attained. If all that is required is to attain low Mars orbit, which would still give the NIMF complete global mobility, then a 2000 K propellant temperature is sufficient. Since this requirement is well below the expected range of performance, it seems clear that the development of NIMF vehicles capable of enabling full global

58

Islands in the Sky

•

16 14

12

.2

-; a: Ill Ill Ill

== QI

u :c

10

8 6

QI

>

2 0

J600

2000

2400

Propellant Temperature

2800

320C

(°K)

Figure 3.8. Performance of the NIMF as a function of temperatur~·-

access on Mars is an objective attainable within the limits of conservative engineering practice. CO 2 can be liquefied out of the Martian atmosphere by simple pump compression at an energy cost of about 84 kWe-hrs/tonne. What this means is that the NIMF, using a 25-kWe power source (either dynamic isotope power source [DIPS], deployable solar, or dual-mode NTR) can completely fuel itself in less than 50 days, without any dependence on surface infrastructure.

BUILDING A BASE ON MARS Once NIMFvehicles have been developed and put into operation, there is no longer an incentive to land the Mars Direct hab modules in widely dispersed locations. Instead, landings can be concentrated at a single location chosen as a good site for a base during the Phase I period of surface exploration. Such a base will support the comparatively large population required to investigate and develop the industrial and agricultural processes needed to create a Martian civilization. While such research in extraterrestrial engineering and resource utilization is proceeding at the base, the field exploration

Mars Direct

•

59

of Mars can be continued at an accelerated pace by parties of scientists making long distance sorties in the NIMF. In Figure 3.9, we see a depiction of a group of habs landed in the same area heing moved together to form a large contiguous pressurized volume. The habs' landing gear can articulate both in the vertical and horizontal directions, allowing them to "walk" the short dbtances required for hook-up. Let us examine how such a base might be built up on Mars using the Mars Direct space transportation infrastructure. We assume that a period of Phase I-type exploration has occurred, with minibases consisting of single four-man hab modules enplaced in a number of locations, and that one of these sites has now been selected for buildup into a major hase. The results are shown in Table 3.3. In Table 3. 3, dates are given in months and years starting from the launching of the team that founds the base. (The "-" mission refers to the Phase I expedition that initially explored the chosen SI•t e. ) Launc hcs arc given • as "h ," " e, " or " c, " f,or ha b, ERV, or cargo launches, respectively, with lower case being used to describe a Phase I (chemical propulsion) type payload, and upper case for Phase II (Ares with NTR third stage). The cargo column is a record of the cumulative amount of useful cargo landed at the base site.

Figure 3.9. Mars Direct habs being mated together to form a large base. (Drawing by Caner Emmart.)

60

•

Islandsin the Sky

Table3.3. MarsDirectbasebuildupplan

Vol./person Date

Crew

1/1 3/3 5/5 7/7 9/9 11/11 1/14 3/16 5/18 7/20 9/22 11/24

4 8 16 16 16 20 24 28 32 36 40 44 48

Launches Habs

lh,le 2h,2e,lc 2h,2e,lc 2h,2e,lc 2h,2e,lc 2H,2E 2H,2E 2H,2E 2H,2E 2H,3E 2H,3E 2H,3E 2H,3E

1 3 5 7 9 11

13 15 17 19 21 23 25

Decks

1 3 7 11

15 21 27 33 39 45 51 57 63

(mJ)

32 48 56 88 120 134 144 151 156 160 163 166 168

Power (kWe)

100 300 500 700 900 1300 1700 3400 5100 10200 12750 15300 17850

Cargo (tonnes)

15 75 135 195 255 385 515 645 775 935 1095 1255 1415

The base buildup strategy is as follows. First a Phase I mission consisting of one hab and one ERV characterizes the site, staying on the surface and exploring the surrounding region for 1.5 years before returning to Earth. The site having been thus selected for developmenton the basisof intimate knowledge of its characteristics, the 1/1 mission consisting of two standard Phase I habs ( one hab deck, one cargo deck), plus two standard ERVs and one,way cargo launch is sent out. The eight,person crew of this mission will stay on the surface for 4 years, departing Mars on 8/5. Such long stays offer the advantage of increasing the total man days spent on the Martian surface at any given launch rate by a factor of two, as well as allowing each base crew to overlap its stay for 1.8 years with the team following on the next launch opportunity. The second team departs Earth on 3/3, arriving at Mars on 10/3. This crew of 8 added to the 8 of the 1/1 team brings the base complement up to 16. Between the Phase I mission, the lower decks of the two habs of the 1/1 mission, and the two cargo launches ( 1/1 and 3/3 ), there are now five cargo decks/workshops at the base. This is quite sufficientfor workshop space, and so the Phase I habs

Mars Direct

•

61

used on the 3/3, 5/5, and 7/7 missions are modified to accommodate habitation on both decks, thus giving the base a rapid expansion of living space per person, as shown in the volume/person column of Table 3.2. Starting with the 9/9 mission, Phase II Ares launch vehicles with an NTR third stage are introduced, increasing the trans-Mars injection throw from 4 7 to 75 metric tonnes. This allows us to expand the hab to three decks, as well as ship about 25 extra tons of cargo with each hab or ERV launch. This eliminates the need for a separate cargo flight and enables the rapid buildup of enplaced base hardware, as can be seen in the landed cargo column of Table 3.2. While the three deck habs of Phase II can accommodate crews of 12, the 9/9 and 11/11 missions only fly with crews of 6 per hab, going to 8 in 1/14, 10 in 5/18, and 12 in 9/22, in order to increase the base crew size at a rate commensurate with the expansion of living space. The 9/9 and 11/11 missions simply send out two 100-kWe reactors of the Phase I type with each ERV launch, but starting with the 1/14 mission, we assume that the reactor power system is switched from thermoelectrics to a Brayton cycle, increasing the power of each unit sent out to 850 kWe. One of these is sent with each ERV, leading to a very rapid buildup of base electric power capability. By the twenty-fifth year of the base, almost 18 MWe of power is in place, far, far, more than the 48-person installation needs for life support or propellant manufacture. Rather, this large amount of power will enable large-scale materials processing, to produce materials to construct vast volumes of pressurized agricultural, industrial, and living space. The highly productive, mechanized, energyintensive industry and greenhouse agriculture that these will contain will create the large-scale economic base required to support the flood of immigrants destined for this Martian Plymouth.

THE COLONIZATION OF MARS This proposition being made publike and coming to the scanning of all, it raised many variable opinions amongst men, and caused many fears & doubts amongst themselves. Some, from their reasons and hops conceived, laboured to stirr up & incourage the rest to undertake and prosecute the

62

•

Islands in the Sky

same; others, againe, out of their fears, objected against it & sought to divene from it, aledging many things, and those neither unreasonable nor unprobable; as that it was a great designe, and subjecte to many unconceivable perils & dangers . . . It was answered that all great & honourable actions are accompanied with great difficulties,and must be both enterprised and overcome with answerable courages. Gov. William Bradford

"Of PlimothPlantation," 1621

By the twenty,fifth year of the Mars Direct base building effort, a large prebuilt town, complete with housing, an electric power grid, and manufacturing and agricultural industries can be in place on the Martian surface.The stage is now set for the colonization of Mars. The difficultyof interplanetary travel may make Mars coloniza, tion seem visionary. However, colonization is, by definition, a one, way trip, and it is this fact which makes it possible to transport the large numbers of people that a colony in a new world needs to succeed. In Figure 3.10 we see a twenty,first century reincarnation of the Mayflower,used to transport immigrants to Mars. An Ares variant with a 1000,klb thrust upper stage lifts 145 tonnes to low,Earth orbit, then a NTR stage hurls the two 40,tonne "habcraft" onto a 7,month trajectory to Mars. After separation from the NTR, the two habcraft separate, rendezvous, and tether off to each other to create Mars's normal artificial gravity at a spin rate of l rpm with the aid of a 700,meter long cable. Arriving at Mars, the tether is dropped, and the habcraft aerobrake, parachute, and land separately on their own sets of hydrogen/oxygen engines. Each habcraft is 10 meters in diameter and includes two complete habitation decks, for a total living area of 157 m2 ( 1700 ft2), allowing it to comfortably house 12 people in space and on Mars. Expansion area is available in the third (uppermost) deck after the cargo it contains is unloaded upon arrival. Thus, in a single booster launch, 24 people, complete with their housing, tools, and inflatable greenhouse, can be transported one way from Earth to Mars. Now let us assume that starting in base year 25, or 2030 A.D., an average of four such boosters are launched every year from Earth

Mars Direct

•

63

figure 3.10. The Martian Mayflower, capable of transporting 24 colonists one way to the red planet.

(perhaps one from each of the four major spacefaring powers, although frankly an effort of this magnitude would be well within the resources of the United States alone). If we then make various reasonable demographic assumptions, the population curve for Mars

64

•

Islands in the Sky

can be computed. The results are shown in Figure 3 .11. Examining the graph, we see that even with this minimal level of effort (and the technology frozen at ultraconservative late twentieth,century levels forever), the rate of human population growth of Mars in the twenty,first century closely approximates that experienced by colonial America in the seventeenth century. This in itself is a very significant result. What it means is that the distance to Mars and the transportation challenge that it implies is not a major obstacle to the creation of a human civilization on the red planet. Rather, the key questions become those of resource utilization, growing food, building housing, and manufacturing all sorts of useful goods on the surface of Mars. Can this be done? We believe so. Unlike the Moon, Mars is rich in carbon, nitrogen, and hydrogen, the key elements needed to support life. Bring a few packets of seeds, several breeding pairs of rabbits and goats, and some inflatable greenhouses to Mars, and a self,expandingagriculture can be established. Plastics, metals, glasses, brick, and concrete can all be manufactured out of native materials (many processes for doing so have been identified and presented in America Mars

1630 1650 1&10 1690 1110 1730 2050 2070 2090 2110 2130 2150 10 I

20

40

60

80

100

120

11so ,no 2170 2190

140

160

YearsAfter Colony Founding

Figure 3.11, Colonization of Mars compared to North America. Analysis assumes I 00 immigrants/yearstarting in 2030,increasingat 2% annual rate, 50/50 male/female. All immigrants are between ages 20 and 40. Average of 3.5 children to an ideal Martian family. Mortality rates are 0.1% per year between ages O and 59, I% between ages 60 and 79, and 10% per year for those over 80.

Mars Direct

•

65

the proceedings of the Case-for-Mars 9.1o,11 conferences) and used to construct large-scale greenhouses, housing complexes, and even environmentally protected towns. As discussed previously, liquid fuels can be synthesized to propel high-powered ground vehicles and construction_ equipment. The required expanding consumer electric power grid can be provided by importing additional nuclear reactors at a rate of 1 850-kWe (dynamic cycle augmented) unit every 4 years, to start, with larger multiunits supplanting them later on in the program. This latter requirement to import nuclear reactors brings up an important question. That is, while it is clear that a Mars colony could manufacture the large majority of its bulk requirements (consumables, propellants, housing materials), it seems equally clear that for a very long time there will remain a necessity to import a certain amount of high-tech manufactured goods from Earth. The question is, is there anything that a Mars colony could export to Earth to pay for such imports, or must Mars always be dependent upon terran charity? One answer is that the Martians could export knowledge. Mars colony could be set up as a university town, with intensive research going on to produce patents which could be licensed for profit to Earthside industries. This may well suffice; however, it is interesting to inquire whether there are any material goods that can be exported from Mars for a profit. At first glance this seems absurd. After all, how many times have we heard that if there were diamonds lying on the surface of the Moon, it wouldn't pay to go there and get them: how much worse must the situation be for distant Mars? Actually, if there were diamonds lying on the surface of the Moon, it would pay to get them, not by flying individual Apollostyle missions, but by setting up a base with a catapult system and firing off the payloads to Earth. The situation on Mars, despite its long distance, is no worse, because at Mars we have indigenous propellant to work with. Consider for example a large unmanned NIMF, a cargo NIMF if you will, capable of lifting 45 tonnes to Low Mars Orbit (LMO). If a methane/oxygen stage is used to fire the payload from LMO to Earth ( where it aero-enters and is picked up on the ground after a parachute landing), about 40% of this payload in LMO, or 18 tonnes,

66

•

Islands in the Sky

can be useful cargo sent to Earth. Now the 45-tonne object in LMO weighs about 17 tonnes in Mars gravity, so for a desirable stage thrust/weight (T/W) ratio of 0.2 for trans-Earth injection we need about 3.4 tonnes, or 7.5 klb, of engine thrust. The typical T/W of a chemical engine is 40, so this means that for each 18 tonnes of useful cargo transported to Earth, an engine weighing 85 kg is required. Let's say that the tanks, and propellant and stage structure can all be manufactured on Mars, but that the engine must be imported, along with another 35 kg of spare parts to replace high-tech items expended in the course of producing the cargo or launching the NIMF. Thus, 18 tonnes of cargo can be transported to Earth at the cost of 120 kg of required imports. This mass transportation ratio, which we term M, is thus 18,000/120 = 150 for this example. The balance of trade ratio, or B, which determines whether the Mars colony can produce any income, is then given by: B=UMP

(1)

where Pis the ratio of the price per kg of the Mars-produced cargo on Earth divided by the cost per kg (production plus transportation costs) of the Earth-produced items, and U is the fraction of total Martian imports that can be expended to support the export operation (the rest is used to support the colony). To continue with the example, let's say that U = 0.1, and that the cost of terrestrial imports is $ 10,000/kg (which is about what it would be if a 121-tonne to LEO class Ares with an NTR third stage costs $500 million per launch). Then if the Martian-produced goods are worth $1000/kgon Earth, we find: B = UMP = (0.1 )( 150)(0.l) = 1.5, and the Mars colony is producing a 50% profit. What could Mars produce that is worth $1000/kg on Earth? One possibility is minerals. Cordell 12 has shown that the history of the Mars, as we know it today, implies that ore-forming processes have occurred on Mars in the past, over an extended period of the planet's geologic history. Thus, while it must be said that such ores have yet to be discovered, geologic science indicates that Mars may be as richly endowed with mineral ores as was the natural Earth. However, unlike Earth, Mars has not had 4000 years of human civilization acting to scour up all the easiest-to-discover and richest ores, so that

Mars Direct

•

67

there is every reason to believe that Mars may possess an abundance of valuable ores far exceeding what is readily available on Earth today. If this proves to be the case, then there are a large number of items that can potentially be extracted from Mars and exported to Earth for profit, as shown in Figure 3.12. Here we have stuck with our above example for the transportation ratio, assumed $10,000/kg costs for imports to Mars, current mineral prices, and examined a range of U values. It can be seen that, while there is considerable uncertainty in the analysis, if minerals at least as valuable as silver or germanium are available on Mars (and there are many such candidates), then a profitable Martian export operation can proceed. It may be added that the introduction of advanced propulsion technology beyond NTR can improve this picture further. For example if a set of solar 13 or magnetic 14 sails were placed in operation cycling between LMO and a trans-Earth injection orbit, cargo delivery could be accomplished solely by operation of these craft in conjunction with the cargo NIMF. The need to export methane/ oxygen engines in large numbers would thus be eliminated from our previous example, and the mass transportation ratio, M, would be increased from 150 to 514, with a very large benefit for the Martian economy. 1000---------------------,

•

'a

100

... ~

-

u ■0.4

-

u ■0.2

-

u ■0.05

u ■0.1

0

•u

.... •. .. ..

10

C

ID

ID

Gadollnlum

~

:I

.,

Yllrlum

Lanlhenum Sliver Germanium Cerium Halnlum

.01 ------------,....------. .01

Rubidium

Galllum Iridium Europium Samarium Platinum

Rhenium

Palladlum

Gold RIIOdlum 10

.I

Price

Rallo

Figure 3.12. Mars base balance of trade. It can be seen that if valuable minerals do exist on Mars, their exploitation with large profit margins is possible.

68

•

Islands in the Sky

TERRAFORMINGMARS One hundred years after the commencement of this colonization effort the population of Mars will reach 80,000 people, and a sufficient industrial base will have been established for serious terraforming operations to commence. McKay15 has offered two design points for a terraformcd Mars. Basing himself on the near,conclusive evidence that Mars was a warm wet planet for the first several billion years of its history (as shown by dry river channels and other water erosion features) he and his collaborators have proposed recreating the primeval greenhouse effect on Mars by reendowing the planet with the same 2 l-iar ( 1 bar = 14.7 pounds per square inch (psi) pressure) CO 2 atmosphere that it had in its childhood. The problem with this proposal is that while 2 bars of CO 2 are required to produce a Mars where the average equatorial temperatures are above the freezing point of water, only about 0.3 bar of volatile CO 2 is actually believed to be currently present on Mars (in the atmosphere, as south polar dry ice, and absorbed in the soil), the rest having become chemically bonded to form carbonates. McKay's second proposal is to endow the planet with a 1 bar nitrogen/oxygen atmosphere. Since these gases are not major green, house effect contributors, in this scenario he proposes supplementing the atmosphere with a "mixed cocktail" of selected artificial CFC gases which have greenhouse potential several orders of magnitude higher than CO The CFC gases, CBrF (halon), C F CFiCl, and CF 2Cl 2 would all be stable in the Martian atmosphere with half, lives of over 200 years, and would not be toxic to humans at the ppm levels required. A production rate of about 100,000 tonnes per year of this mixture would be required to sustain a greenhouse effect on Mars sufficient to raise the equatorial region's temperatures to Earth temperate zone levels (currently about 650,000 tonnes of these gases are produced per year on Earth). The estimated power required to support the steady state CFC manufacturing on Mars to maintain this greenhouse would be about 500 MWe, although a production of 1,000,000 tonnes/year (5000 MWe) would be required if the greenhouse is to be established within a 40,year time span. These requirements hardly seem excessive for a project of this value. The problem, however, is that while the CFC greenhouse appears techni, 16

•

2

,

3

2

6

Mars Direct

•

69

cally feasible, the nitrogen/oxygen atmosphere does not, as chemi, cally free nitrogen is not to be found in anything like the required quantities on Mars. Given this situation, our proposal is for a compromise combining what is most feasihle from the two McKay scenarios. That is, what we propose is to set up the CFC production on Mars and start to greenhouse the planet despite the fact that no nitrogen/oxygen atmosphere is to he had. The first thing that will happen as the planet starts to warm is that the CO 2 as dry ice in the south polar ice cap will vapuri:e, thickening the CO 2 atmosphere, and adding to the greenhouse effect of the CFCs. Within 40 years, nearly all of the volatile CO: can he made to outgas, forming an atmosphere of about 0.3 har pressure ( 4.4 psi) on Mars. While this atmosphere would not be hreathahle by humans, it would eliminate the need to wear pressure suits. Rather, humans working on the outside on Mars would be able to wear street clothing, along with a kind of scuba gear breathing a mixture of 3 psi oxygen and 1.4 psi nitrogen out of a tank. As temperatures rise, the water frozen as permafrost into the Martian soil will begin. to melt and flow again into the long dry river channels of Mars. Soon the rains would begin to fall, and it will become possible to grow plants in the open across large regions of the planet. Within 100 years of the commencement of the terra, forming effort, desert areas such as Chryse Planitia may be covered with amber waves of genetically modified grain, while the lower slopes of the Olympus Mons are decorated with proud stands of ponderosa pine. These planes will do more than produce food and look grand; they will transform Mars's atmosphere, just as their ancestors did for the Earth. The atmosphere they create will be nearly pure oxygen, at about the 3 psi level. As the experience of the Apollo and Skylab programs prove, humans can function quite well for extended periods of time in such an atmosphere. Thus, with the help of our biospheric allies, the day may come when humans standing on the surface of Mars doff their breathing gear and gratefully breathe in the cold crisp air of the Martian spring. In 500 years, a space of time equal to that which separates the present age from that of Columbus, a civilization of over a billion people may live on a new Mars, a world not discovered, but created by the courage and enterprise of the human spirit.

70

•

Islands in the Sky

Acknowledgments The CAD drawings used in this article were done by Robert Spencer, while the paintings were done by Robert Murray. The sketch of the Mars base being assembled was done by Carter Emmart. Performance trajectory analysis on the Ares launch vehicle was done by Sid Early, while many useful engineering suggestions were made by Jim Greenwood. Dr. Chris McKay provided essential background in terraforming science. Dr. Ben Clark provided a valuable critique of the basic Mars Direct concept, and Al Schallenmuller has provided corporate support and assistance with the analysis of the methane/ oxygen production process.

Chapter 4 Stephen L. Gillett

Inward Ho!

Mercury's the ugly duckling of the inner system; it's been pretty much ignored hnrh in SF and in space development scenarios. "On to Mars!" "Out to the asteroids!" have been the slogans. "In to Mercury!" just doesn't have a lot of pizazz. Charles Pellegrino and James R. Powell have bucked this trend a little; in Analog, 1 they proposed covering Mercury with a solar-cell farm to collect the vast power required for interstellar flight. But there's more to Mercury than that! Don't cover over the planet itself; we need it for other things. Build the collector farm off-planet instead, as Robert L. Forward suggested in Flight of the Dragonfly. 2 'Cause Mercury's got a lot of potential value, and not just because of the solar power there. So let me digress on a synergy between highefficiency solar sails, Mercury, space development, and terraforming.

SOLAR SAILING AND MERCURY Light boats sail swift . . . Shakespeare

Troilusand Cressida,Act IP

71

72 •

Islandsin the Sky

As a solar sail gets real thin, it can support its own weightand then some-while at rest with respect to the Sun. Obviously such an "ultrasail11 is more efficient than a sail that can't support its weight. But its advantages run deeper than that. For one thing, as in the story "SunStat11 by Oltion and Goodloe, 4 an ultrasail can hold a platform at rest with respect to the Sun ( or another star, for that matter), which is convenient for many sorts of observations. But more than that, the closer to the star, the better the acceleration-despite being deeper in the gravity well! Explanations follow. For a perfectlyreflectingsail, the pressure of sunlight is given by: P

= 2S/c,

where P is the pressure, S the solar constant, and c the speed of light. The solar constant is the amount of energy that a given area receivesfrom the Sun in a certain time, and has units of power per area (W/m2, say). For a perfect absorber, the pressure is exactly half the above:

P = S/c. At the distance of Earth, the value of solar constant is about 1340 W/m2• Using c = 300,000 km/s, we get a pressure of 4.5 X 106N/m2, or 6.5 X 1010 psi, for a perfect reflector. Not a lot! (The totalforce, of course, will depend on the total area of the sail.) Now, for various materials we can calculate the critical thickness at which a sail would be self-supportingat the distance of Earth. To do this, we need the density and reflectivity of the material, and the gravitational field of the Sun. Such data are in standard tables. For aluminum (not a perfect reflector), the critical thickness works out to be about 0.25 µ.m (i.e., microns-millionths of a meter), or about two-hundredth the thickness of a human hair. (The actual calculation is left as an exercise for the reader!) Obviously, a useful sail must be even thinner than this, to have any capacity to hold a payload. Furthermore, even though l specified"at the distance of Earth," it turns out this thickness is independent of distance. For con-

Inward Ho!

•

73

sider: If a sail is in balance, the outward force of the sunlight on the sail and its burden must equal the inward pull of gravity. Now mentally move this sail inward. The attractive force on it increases substantially, according to Newton's law of gravitation; it goes up as the inverse square of the distance. But the sunlight pressure also increases substantially. In fact, it increases in exactly the same inverse-square proportion, because light intensity drops off as the inverse square of the distance. Thus the effects exactly cancel. So if the sail hovers at the distance of Earth, it hovers anywhere. Now consider a sail that can support more than its own weight plus payload. It cancels out gravity and then some, so there's a small net pressure left. Then what happens when we bring the sail sunward? Again, the Sun's gravity and sunlight intensity both increase according to the inverse-square law, so the closer you are, the more the sunlight pressure exceeds the gravitational pull. For example, say you were at a distance of 1 AU (astronomical unit-the average distance from Sun to Earth) and the light pressure on your ultrasail is 1. 1 times its weight from gravity. Now move in to 1/2 AU. Gravity is now four times greater, but the light pressure is also four times greater. Now, 4 X 1 is 4, but 4 X 1. 1 is 4.4, so the net force outward on the sail has also increased four times! The upshot is that the excess pressure on the sail acts like an inverse-square repulsion. Obviously, then, the closer to the Sun, the stronger the force, and the better the acceleration. This has some bizarre consequences. For example, if you were going from Earth to Pluto with an ultrasail, it's far faster to drop in toward the Sun first. The extra acceleration near the Sun much more than compensates for the longer distance. (To calculate travel times from point A to point B in an inverse-square repulsive field, you can either integrate the inverse-square potential from first principles, or else look up the equation-which is much faster. I'm not going to give it here; it's given in reference 5. The equation is kind of messy, but that's what computers are for.) So the closer to the Sun the better. Now, what's the closest planet to the Sun in the system? Right! Mercury is one dandy place to start ultrasails from.

74

•

Islands in the Sky

MERCURY AND RESOURCES Mercury-A Mini-Earth in Moon's Clothing? Murray, Malin, and Greeley 6

Mercury is probably one dandy place to make solar sails, too. Not only is it routinely the closest piece of matter to the Sun (and far and away the biggestpiece, too), but it probably has an aluminumrich crust. Why? I have to digress again, this time for a review of the state-of-the-art about Mercury. As everyone knows, we found out in 1964 that the one thing we knew about Mercury wasn't true. It does not keep the same face to the Sun, but instead rotates with a 3/2 resonance (3 rotations in 2 revolutions). The error came about because the planet's very difficult to study from Earth. Although it's bright ( l 've seen Mercury, a sight Copernicus is said to have missed) it's always close to the horizon, visible only briefly after sunset or before sunrise, while the sky is still twilit. Now, with its true rotation, you see Mercury facing the same way every other orbit; but since it was very difficult to see Mercuryat all, observers just saw Mercury facing the same way many times. They then naturally assumed it must always face the same way, and thus must be facing the Sun. The physicists have since shown that the 3/2 resonance is much more stable than the 1/1 resonance (Mercury always facing the Sun), and so we should have known beforehand there was a problem; but this sort of post,hoc analysis is not terribly impressive! If from Earth you can't even be sure of the rotation period, you can figurethat Earth-based telescopic studies are pretty useless. Thus, all detailed info on the planet comes from the Mariner 10 fly,by back in 1974. This mission was originally intended as a single fly, by, with the spacecraft then continuing on in an independent orbit around the Sun. Fortunately, someone thought to calculate where Mercury would be when the spacecraft crossed its orbit again, and sure enough, Mercury would be there again, too! So Mariner 10 made three separate passes by the planet and took pictures each time, before it ran out of attitude-control fuel. Unfortunately, due to Mercury's slow rotation and the details of orbital mechanics, the spacecraft saw the same view each pass, so one Mercurian hemisphere

Inward Ho!

•

75

is still unknown. At this point, though, there's no reason to think it's spectacularly different from the side we've seen. The first impression from the Mariner photos is that Mercury looks like the Moon. It's pocked with craters upon craters, very much like the lunar highlands. And also like the lunar highlands, it includes some very large craters indeed-"basins," as the planetary scientists now term them, hundreds to thousands of kilometers across. We know from dating Apollo samples chat the "late heavy bombardment" recorded on the Moon had essentially ceased by 3.9 billion years ago. If the massive cratering on Mercury reflects the same bombardment, which seems reasonable, Mercury is a pretty inactive place. No other surface processes have modified that ancient craterpocked crust, other than a little bit of "faulting" ( i.e., fracturing), probably reflecting slight contraction of the Mercurian core, and some subsequent desultory cratering. There's also no evidence of any ancient atmosphere; as Murray et al. 6 note, crater rays-bright streaks centered on the crater, consisting of a thin layer of powdered rock thrown out by impact-are not only common but preserved. It takes very little atmosphere to blow them away, or to keep them from being deposited in the first place. So why is it like a "mini-Earth"? Mercury has no air, no water; it is inactive, unchumed by plate tectonics, unmarred by erupting volcanoes . . . it hardly seems Earthlike! It is "Earthlike" in its density. Mercury has a density of 5.42 grams per cubic centimeter (g/cc), little less than Earth's 5.52 g/cc. (The Moon's average density, by comparison, is only 3.3 g/cc, not much more than ordinary rock.) The high density indicates that Mercury must contain a lot of dense material, material much denser than silicate rocks. A very un-Moonlike characteristic. Earth's high mean density results from its iron-nickel core. And similarly for Mercury; it has a large core. ( It even has a small magnetic field, also in contrast to the Moon.) In fact, Mercury's core makes up about 80% of the mass of the planet. But the core's relevance for near-term resources is not clear. Even this big core is still pretty deep, buried under silicate rocka crust and mantle-500-600 km thick. Digging thar: deep is going to take a while. (There's a little slop in these estimates, since we do not know exactly how dense the core is. Murray et al. note, though, that even with generous estimates of the thickness of the

76 •

Islandsin the Sky

silicate crust and mantle. Mercury's core is bigger than Earth's Moon!) So for the time being, resource,wise, we're going to have to settle for what's in the silicate mantle and crust atop that large core. As a working model, they probably are very similar to the Moon's. Based on evidence from rocks collected by the Apollo missions, we now think the Moon was almost or completely molten after its formation-a notion picturesquely termed the "magma ocean." As the magma ocean cooled, minerals started to crystallize in it; most sank, but one in particular floated-plagioclase feldspar, a calcium, aluminum silicate. If things settled out undisturbed like a jar of salad dressingleft in the fridge, you'd end up with a nicely layered planet; a plagioclasecrust outside, and layers of denser, iron,magnesium, rich silicates inside. And to first order, that's a good description of the Moon. The top of that ancient plagioclase crust shows up as the lighter, densely cratered surface of the lunar highlands. (Of course, natural processesare never this simple. For one thing, the Moon's plagioclasecrust is pervasively fractured by those mammoth craters, and bodies of iron,rnagnesiurn silicate rock squeezed into this frac, tured crust, like mud pushing up through a broken sidewalk. For another, some subsequent melting occurred as heat built up in the Moon from the decay of long,lived radioactive elements, and this led to some early volcanic activity-which caused the smooth dark lava plains, or "maria.") So, if the silicate part of Mercury's like the Moon, there's bound to be a big variety of a lot of useful stuff right at the surface, all combined in silicates: oxygen, calcium, magnesium, iron, some so, diurn, and potassium.Rarer elements that have an affinity for oxygen and silicates are also likely to be concentrated locally; useful things like uranium, the rare,earth elements, zirconium, tungsten, . b.1urn.... 7 mo And in particular, aluminum-the raw material for solar sailsis probably one of the most abundant elements at the Mercurian surface, bound up in that feldspar crust. By the same token, however, Mercury is probably essentially volatile,free, like the Moon. So you will have to bring all your water, carbon, and organics. As on the Moon, there are probably some volatiles, especially hydrogen, stuck in the regolith as a result of being implanted by the solar wind (the tenuous stream of atomic

Inward Ho!

•

77

particles continually emitted by the Sun), and they'll help keep life, support stocks replenished, but that's probably all. (Oh, well. You can't have everything.) Anyway, Mercury is also easy to get off. For all intents and purposes, there's no air, so mass drivers work fine. ( If you want to quibble, there's a tenuous "atmosphere" consisting of a steady-state population of helium atoms detained awhile from the solar wind. It's interesting only to particle physicists.) And although Mercury has over twice the escape velocity of the Moon, it's about twice as easy to leave, if you use mass drivers or some other electrically powered technology. Mercury has almost seven times the solar flux as the Moon ( the exact amount varies due to the ellipticity of Mercury's orbit), and that more than compen, sates for the additional gravity. With just modest solar collectors you can get a lot of power. (To be sure, sunrise and sunset are a problem; solar cells don't work well at night. So you set up your fledgling operations near one of the poles; like the Moon, Mercury has little axial tilt, so there you can be in sunlight nearly year-round. And as a power grid spreads across the planet, the day/night cycle will become less and less a problem.) And this echoes the point Pellegrino and Powell made: Mercury has lots of solar power. Its position is in stark contrast with the asteroids, which are a ballyhooed "obvious" destination for space development. Sure, there's lots of stuff out there in the Belt, volatiles and raw metal and whatnot, but there's not a lot of energy out there. From the old inverse-square law again, the ratio of solar flux at Mercury to that at a typical mainbelt asteroid-Ceres, say-is about 160; over two orders of magnitude. Thus, to yield the same energy, solar collectors of the same efficiency must be more than ten times as long on a side (because the square root of 156 is about 12.5). To get the same power, an acre of collector at Mercury blossoms to 160 Ceres. (All right, if you stick with acres-a quarter,section!-at near-Earth asteroids your solar power is more abundant. Near-Earth asteroids are a highly limited population, however, and we're going to be moving afield from there before too long.) Ultimately, perhaps, we might even take Mercury apart. Reach, ing that gigantic nickel-iron core would be a big resource incentive. And, as a geologist myself, dismantling Mercury would have major scientific interest; dissection of a non-living planet would tell us a

78

•

Islands in the Sky

lotmore about how planets work. (Of course, this would not happen for a long time.) So why has Mercury'spotential been so ignored? There's probably something emotionally disturbing about going in closer to the parent star. In part, maybe all that old SF about the hellish day side of Mercuryhas sunken into the collective subconscious, as a fundamen· tally irrational but very real fear. Or, maybe going inward goes against the psychologicalgrain; we're going into space to get away from the confinements of Earth, to broaden our options, and so to move out "toward" other stars-however minutely-is more satisfying. But any worry is psychological. Keeping a Mercury base cool will be no problem for twenty-first century technology, especially since all the actual mining and extraction operations will be robotic. For human habitation, just mound some regolith over the modules and it'll be just like the Moon. Mercury orbit is also an overlooked place for spacecolonies, if that's what you're into. And finally, two possible inner-systemplaces, anticipated by George 0. Smith in his classic "Venus Equilateral" stories, are the Venus L-4 and L-5 points. There may even be some asteroidal debris-"Venus Trojans"accumulated there, to bootstrap the Mercury operations.

MERCURY:THE KEY TO VENUS? A curious shiny meteor eased into the planet's soupy atmosphere. As it plunged, atmospheric resistance heated it to glowing; then sudJenly, as it slowedinto the denser layersbelow, it ignited. Thick black smoke trailed the burning meteor as it fell through the thick CO 2 atmosphere. . • •

As Analogreaders may remember, I discussed terraforming Venus in two previous articles (Dec. 1984 and Nov. 1985). At the Eighteenth Lunar and Planetary Science Conference-the "LPSC," held at the Johnson Space Center, Houston, in March, 1987-J im Oberg put together an informal conference called Terraforming II, and invited interested people to come give talks. So I did. (Alas, although the LPSC lent us a room, T erraforming II was not part of its formal proceedings. T erraforming's still too far out. However, progress is being made. Back around 1976, another associated meeting, on "Lunar Utilization," also was not a formal part of the LPSC because

Inward Ho! •

79

it was too far out. In 1987, though, the LPSC had a formal session on "Space Utilization"!) Anyway, at Terraforming II, I proposed a new notion for terra, forming Venus, inspired by Freeman Dyson's Enceladus Project, a "thought experiment" for terraforming Mars. 8 To briefly summarize Oyson's scenario: A robot package is sent to Enceladus, a large Saturnian moon assumed (for the sake of argument) to consist mostly of water ice. The package differentiates into several types of self-replicating robots, which eventually start launching packages of ice to Mars, guided by solar sails. As a steady stream of ice meteoroids warms the Martian atmosphere, water col, leccs, rain falls, and volia-a habitable planet. But what about Venus? As I reviewed in my December 1984 article, the overwhelming problem with terraforming Venus is getting 90 atm worth. All that rid of its massive CO 2 atmosphere-about air keeps the surface near red heat because of a powerful greenhouse effect; the atmosphere acts like a blanket, letting sunlight in but hindering the escape of heat. And till the CO 2 is gotten rid of, sending a stream of Enceladus ice to Venus \Votdd just make things hotter. Water vapor itself is a dandy greenhouse gas. In fact, Venus probably arrived at its present sorry state through the so-called "runaway greenhouse"; if a planet's surface hecomes too hot, enough water vapor evaporates so the greenhouse effect gets more efficient, so things get even hotter, so yet more water evaporates. . . . The oceans boil soon, and for a while you have an atmosphere of steam. (Quickly, as geologists measure time, the steam atmosphere is destroyed by solar ultraviolet, which breaks up water molecules so the hydrogen can escape to space.) Earth has just as much CO 2 as Venus, but conveniently, it's virtually all locked away in carbonates in the crust, as limestone, mainly. Venus never had oceans (or lost them early on), so carbonates were never stable, so the CO 2 stayed in the air, so things got real you were going to ask-we can't react Venus's And-if hot .... air with its crust now; you'd have to garden the crust much too deeply, even if you could get quick and complete reaction. (And you'd get neither; the silicates making up rocks react too sluggishly.) In addition, things would need to cool for the rocks to react, but they can't cool till the COz's been removed! A classic Catch-22.

80 •

Islandsin the Sky

Seeding the Venus atmosphere with algae to break down the CO2-Carl Sagan's venerable terraforming suggestion, back in the 1960s-won't work either, at least for the whole job. There's just too much air. Breaking up 90 atm of CO 2 gives you over 60 atm of oxygen, plus a layer of carbon over 100 m thick. They won't stay separated long; C + 0 2 =CO 2.Boom! We need to get rid of the air. Well, can we remove the CO 2 bodily? That's very hard. We could maybe seed the planet with self,rep robots to gather up the atmosphere (which involves compressing it, perhaps a kilo or so at a time) and shoot it off into space. But Venus is a big planet; you need a lot of !::t. V to escape her, and because of her atmosphere, acceleration is difficult;catapults like mass drivers don't work because of air resistance. What about precipitating out the CO 2 by reacting it with some, thing from outside? Ah! Maybe here's the key. The best reactant would be calcium or magnesium metal. Both will tend to reduce carbon dioxide to carbon, stripping out the oxygen: 2 Mg+ CO 2 = 2Mg0 + C and 2 Ca + CO 2 = 2CaO + C. Magnesium actually bums in CO 2. It's a flashy lab demonstration; just poke a burning piece of magnesium ribbon into a bell jar full of CO2 and watch the black smoke of carbon boiling off. With both calcium and magnesium,'moreover, the oxides further react with more CO 2 to make carbonate: MgO + CO2 = MgCO 3 and similarly for calcium. So, each atom of metal ends up removing a molecule and a half of CO 2. Not bad. The problem? Not surprisingly,neither calcium nor magnesium occur uncombined in nature. They are reactive, after all. So the metal would have to be extracted, and that's hard; magnesium and

Inward Ho!

•

81

calcium are both tightly bound into silicates and require a lot of energy to be broken out. And, finally, a preposterous amount of metal is required, about 8 X 1020 kg calcium or 5 X 1020 kg magnesium. To give you a feel for how preposterous, this is approximately the mass of Ceres, the largest main,belt asteroid. If all this metal was formed into a sphere, it would be something over 1000 km in diameter. It's absurd to think we could ever refine that much raw metal, especially for an economically marginal proposition like terraforming Venus. Just another silly science,fiction fantasy! Or is it? Back to Mercury. {You thought I'd forgotten Mercury?) If its crust is really like the Moon's, it is fundamentally madeof calcium, aluminum feldspar, abundantly invaded with magnesium,rich igne, ous rocks. So both calcium and magnesium are among the most abundant elements at its surface. Now, as I said, silicates are hard to break up, especially without a lab full of reagents. But it can be done. One simple (that's key!) way is to melt the rock. You can electrolyze it; stick electrodes into each side of your pool of molten rock, hook them up to a big battery, and watch. It's just like breaking water up into hydrogen and oxygen. Oxygen will bubble off the anode, and a mixture of metals will dribble off the cathode. And the melt's resistance to current provides the heating to keepit melted. Well, of course it's not that simple. Silicate melts are very hot (over l000°C, in this case), which makes them tricky to deal with. You're dealing with lava, after all. ( It's easier in a vacuum, anyway, because they're easier to insulate.) They're corrosive; things like electrodes tend to dissolve. Preventing undesired side reactions at the cathode is another sticky issue: Only the least reactive metal tends to separate, since any more reactive metal tries to displace a less reactive metal from the melt instead. (The anode also tends to plate up with gunk for similar reasons.) And we've ignored the question of collecting the dirt and melting it in the first place. But none of these problems is fundamental. (To quote a fatuous phrase, "they're just engineering!") And refining silicates in this way has lots of advantages: You don't need crushing or sorting; your feedstock is whatever random bunch of regolith you scoop up. You don't need water, with an expensive and complicated wet,chemical extraction facility; your working fluid is just molten rock. And finally, some work is already being done on such processes, to determine

82

•

Islands in the Sky

their feasibility for lunar smelting. The technology will be available when we need it. So, let's extrapolate. From what's learned from the Enceladus Project and a couple generations of Moon-mining, the Venus Terraforming Project builds a package and sends it to Mercury, where it differentiates into several types of self-replicating robots. Some scoop up dirt, some set up solar-power collector and power distribution networks, some set up electrolysis vats, some build mass drivers. • • • And they start electrolyzing regolith into raw metal, and fashioning it into chunks, each with a little solar sail and a guidance chip telling it how to get to Venus. Soon, the auto-assembled mass drivers start throwing the chunks off the planet, where their solar sails deploy. The guidance chip then steers the way to Venus, where the ingot smacks itself into her air, a rain of artificial meteors each munching up another kilogram or two of CO 2. Eventually the job is done. Of course, once the atmosphere's thinned down quite a bit we'll need to add some Water, but that's trivialby comparison; toward the end of the project, when there's no danger of retriggering a runaway greenhouse, the Enceladus project will send some ice cubes to Venus as well. There's some curious economics here, hinted at by Dyson in his thought experiment. Each chunk of metal is nothing more than a self-guided ingot (self-guided with exquisite accuracy), and trillions of such lumps of metal, each with a tiny sophisticated guidance system, will be sacrificed utterly to atmospheric incineration. Bizarre! Not a way to treat precision machinery. But the destruction of that exquisitely organized ingot does not matter; what matters is not organization per se but how cheaply it can be made. And the cost is virtually zero. It's analogous to products from living things on Earth. For example, consider a piece of firewood. Look at the exquisite pattern of this wondrous stick, the phenomenal organization of thousands of cells-each itself a marvel of complexity-that generated it. And you're going to casually burn it for fuel? But of course. We think of wood as a "natural" product, easily made and casually destroyed, and so it is. Biological self-replicators produce it cheaply as a byproduct of their existence. Or, consider an apple. It's just as marvelous as the firewood; but in addition, each one of those seeds, so thoughtlessly discarded, is a package that can build a whole new tree! But again, apples are

InwardHo! •

83

made by the million; ironically, many fewer would be made if they couldn't be destroyed for food. Who's going to raise an orchard of fruitless trees? Or consider a krill-eating whale, filtering tiny organisms out of seawater by the ton. Each one of those zillion planktonic lifeforms is a marvel of complexity unto itself, a whole independent life form, but each is doomed to he disaggregated and reformed into whale-stuff. The degree of organization is not important. The important thing is how cheaply you can organize. Given self-rep robots, their products are no more valuable, and no less valuable, than the products of living things. That is, they are valuable for what they are, with no value added for the organization that went into them. How long might terraforming take? The lower limit is set by how much solar energy it takes to make all that metal and send it to Venus. There are two main costs: (1) busting it out of the rocks, and (2) sending it out of Mercury's gravity well. For magnesium, the minimum extraction energy is going to be about 25 kilojoules per gram (kJ/g); for calcium, about 18 kJ/g. And then it takes about 9kJ/g to get something off Mercury. In both cases, the total energy we need adds up to about that in all the sunlight intercepted by Mercury in 4 years! And this assumes best-case extraction efficiencies; no energy lost, and completely pure silicate feedstock, so for real-world efficiencies the time goes up accordingly. (Of course, we could help things along by gathering additional solar power with off-planet collectors.) And how much Mercury did we chew up to send to Venus? A lot. If we figure out how much rock we had to process to get that much metal, we arrive at something around 3 to 5% of the mass of Earth's Moon! This is not trivial; a great deal of Mercury's surface will be affected. But this is not necessarily bad, either. By building extensive observational and recording capability into the self-rep robots, we could get unparalleled data on the internal constitution of Mercury as it was excavated. In fact, economics being what they are, probably much more scientific data could be gotten by piggy-backing onto such a project than could be obtained from a purely scientific investigation. (This also holds for the Enceladus project.) And, if we're clever, at least one of the excavations could reach that iron-rich core, with all the scientific and resource potential that promises.

84

•

Islands in the Sky

In fact, granting such a self-differentiating, self-replicating system, it turns out the big constraint is not how much metal you need but how much energy you're adding to the Venus atmosphere. If you're not careful, a tremendous amount of matter would be falling in from essentially infinity, and all that gravitational energy will show up as heat. My BOTE" calculations suggest the heat imparted in that case is about 100 times what's in the atmosphere already! (And, of course, one of the things we must do is cool Venus, not cook her.) To avoid this heat influx, you need some tricky guidance to decelerate before hitting the atmosphere. (And the deceleration can't involve atmospheric braking, because-after all!-that just dumps energy into the atmosphere.) So. Would it work? I think so. Is it the best way? How should I know! And why terraform Venus anyway? That's a different question ... but the reason not to terraform Venus is not "because it's impossible!" (Dr. Bob Forward, well-known to Analog readers, also attended the Terraforming II conference. He gave the talk after mine. After I had finished, he said he thought he was supposed to give the sciencefiction scenarios. I took it as a compliment!)

• Back-of-the-envelope

Chapter 5 Rohert M. Zubrin

Colonizing the Outer Solar System

Currently, the outer solar system is largely a curiosity. Despite its immensity, its four spectacular giant planets, its mysterious minor planet, six moons of planetary size, and scores of smaller moons, as well as several known and probable myriads of unknown asteroidal and cometary objects of every description imaginable, the distance separating us from this vast domain appears to relegate it to merely academic interest. Such appearances are deceptive. It is safe to say that within a century, a mere blink of an eye in human history, it will be evident that these outer worlds contain the keys to continued human survival and progress, to war and peace, and to humanity's hopes for the stars. Let's see why.

SOURCES OF POWER To get a glimpse of the probable nature of humans in the late twenty, first century, it is first necessary to look at the trends of the past. The history of the technological advance of the human species can be written as a history of increasing utilization of energy. Consider 85

86

•

Islands in the Sky

the energy consumed not only in daily life, but also in transportation and the production of industrial and agricultural goods. Americans in the electrified 1990suse approximately three times as much energy per capita as their predecessorsof the steam and gaslight 1890s, who in turn had nearly triple the per capita energy consumption of the pre-industrial 1790s. Some people see this trend as a direct threat to the world's resources,but such rising levels of energy consumption historically have correlated directly with rising living standards. If we compare living standards and per capita energy consumption of the advanced sector nations with those of the impoverished Third World, we see that they are still directly correlated. This relationship between energy consumption and the wealth of nations will place extreme demands on humanity's present set of available resources. First, to raise the entire presentworld population to current American living standards (and in a world of global communications, it is doubtful that any other arrangement will be acceptable in the long run) would require increasing the global energy consumption at least a factor of ten. However, world population is increasing, and while global industrializationwill slow this trend, it is probable that population levels will at least triple before they stabilize. Second, current American living standards and technology utilization are hardly likely to be the ultimate (even in late twentieth-century America, there is still an abundance of poverty) and will be no more acceptable to our descendants a century hence than those of a century ago are to us today. All in all, it is clear that the exponential rise in human energy utilization will continue. In 1992 humanity consumed about 12 TW (TW = terawatt = 1 million MW) of power; at the current 2.6% growth rate, we will be using nearly 200 TW by the year 2100. The total anticipated power utilization and the cumulative resources used (beginning with 1992) are given in Table 5.1. For comparison, the total known or estimated energy resources are given in Table 5.2. In Table 5.2, the amount of He3 listed for each of the giant planets is that present in their atmospheres down to a depth where the pressure is ten times that of sea level on Earth. If one extracted at a depth where the pressure is greater than this, the total available He3 would increase proportionally. If we compare the energy needs for a growing human civilization with the availability of resources, it is clear that, even if the environmental problems associated with

Colonizing the Outer Solar System

•

87

Table5.1. Projected human use of energy resources Year

Power (TW)

Energy Used after 1992 (TW-years)

2000

15

107

2025

28

623

2050

53

1600

2075

101

3460

2100

192

7000

2125

365

13700

2150

693

26400

2175

1320

2200

2500

50600 96500

burning fossil fuels and nuclear fission are completely ignored, within a few centuries the energy resources of the Earth and Moon will be effectively exhausted. Large-scale use of solar power can alter this picture somewhat, but eventually the enormous energy reserves in the atmospheres of these giant planets must and will be brought into play. Thermonuclear fusion reactors work by using magnetic fields to confine a plasma consisting of ultra-hot charged particles within a vacuum chamber where they can collide and react. Because highenergy particles can gradually fight their way out of the magnetic Table5.2. Solar system energy resources Resource

Amount (TW-years)

Known terrestrial fossil fuels

3000

Estimated unknown terrestrial fossil fuels Nuclear fission without breeder reactors

7000

Nuclear fission with breeder reactors Fusion using lunar He3 Fusion using Jupiter He3 Fusion using Saturn He3 Fusion using Uranus He3 Fusion using Neptune He3

300 22,000 10,000 5,600,000,000 3,040,000,000 3,160,000,000 2,100,000,000

88

•

Islands in the Sky

trap, the reactor chamber must be of a certain minimum size to delay the particles' escape long enough for a reaction to occur. This minimum size requirement generally makes fusion power plants unattractive for low-power applications, but in the future, where human energy needs will be on a scale tens or hundreds of times greater than today, fusion will be far and away the cheapest game in town. In the twenty-first century, nuclear fusion using the cleanburning (no radioactive waste) deuterium-helium-3 reaction will be one of humanity's two primary sources of energy, and the outer planets will be the Persian Gulf of the solar system.

THE PERSIANGULF OF THE SOLAR SYSTEM Today, Earth's economy thirsts for oil, which is transported across oceans from the Persian Gulf and Alaska's North Slope by fleets of oil-powered tankers, for use in industrial sectors. In the future, the inhabitants of the inner solar system will have fuel for their fusion reactors delivered from the outer worlds by fleets of spacecraft driven by the same thermonuclear power source. Although the ballistic interplanetary trajectories made possible by chemical or nuclear thermal propulsion are adequate for human exploration of the inner solar system and unmanned probes beyond, something much faster will be needed to sustain interplanetary commerce encompassing the gas giants. Fusion reactors powered by D-He3 are a good candidate for a very advanced spacecraft propulsion system because the fuel has the highest energy-to-massratio of any substance found in nature, and also because in space the vacuum in which the reaction must run can be had for free in any desired size. A rocket engine based on controlled fusion could work by allowing the plasma to leak out of one end of the magnetic trap, adding ordinary hydrogen to the leaked plasma, and then directing the exhaust mixture away from the ship with a magnetic nozzle.The more hydrogen added, the higher the thrust, but the lower the exhaust velocity. For travel to the outer solar system, the exhaust would contain more than 95% ordinary hydrogen, and the exhaust velocity would be over 250 km/s ( which compares quite well with the exhaust velocities of chemical or nu-

Colonizing the Outer Solar System •

89

clear thermal rockets of 4.5 and 9 km/s, respectively). Large nuclearelectric propulsion (NEP) systems using fission reactors and ion engines, a more near-term possibility than fusion, could also achieve an exhaust velocity of 250 km/s. However, because of the complex electric conversion systems required, such NEP engines would weigh about eight times as much as fusion systems, and the trips would take twice as long. If no hydrogen is added, a fusion configuration could theoretically yield exhaust velocities as high as 15,000 km/s, or one-twentieth the speed of light! Although the thrust level of such a pure D-He3 rocket would be too low for in-system travel, the terrific exhaust velocity would enable voyages to nearby stars with flight times of less than a century. Extracting the He3 from the atmospheres of the giant planets will be difficult, but not impossible. A winged transatmospheric vehicle is required that can use the planet's atmosphere for propellant, heating it in a nuclear reactor to produce thrust. I call such a craft a NIFT (Nuclear Indigenous Fueled Transatmospheric vehicle). After leaving its base on one of the planet's moons, a NIFf would either cruise the atmosphere of a gas giant, separating out the He3, or rendezvous in the atmosphere with an aerostat station that had already produced a shipment. In either case, after acquiring its cargo, the NIFf would then fuel itself with liquid hydrogen and then rocket out of the atmosphere to deliver the He3 shipment to an orbiting fusion-powered tanker bound for the inner solar system. In Table 5.3 we show the basic facts that will govern the commerce of He3 from the outer solar system. The flight times listed are one-way from Earth to the planet, with the ballistic flight times being those for minimum-energy orbit transfers. Although these can be shortened somewhat at the expense of propellant (gravity assists Table5.3. Getting around the outer solar system One-Way FlightTime

Pla~t Jupiter Saturn

Uranus Neptune

Disumce from

Ballistic

NEP

Sun (Au)

(:yn)

('Jn)

(::,rs)

Velocicyto Orbit (le.mis)

NIFT Vehicle MassRatio

2.7 6.0 16.0 30.7

2.2 3.0 5.0 6.6

1.1 1.5 2.5 3.3

29.5 14.8 12.6 14.2

23.7 4.6 3.6 4.3

5.2 9.5 19.2 JO.I

Fusion

90

•

Islands in the Sky

can help as well, but are available too infrequently to support regular commerce), they are too long for commercial traffic to Saturn and beyond (even if the vessels are fully automated, time is money). The NEP and fusion trip times shown assume that 40% of the ship's initial mass in Earth orbit is payload, 36% is propellant ( for oneway travel; the ships refuel with local hydrogen at the outer planet), and 24% is engine. Jupiter is much closer than the other giants, but its gravity is so strong that even with the help of its very high equatorial rotational velocity, the velocity required to achieve orbit is an enormous 29.5 km/s. A NIFf is basically a nuclear thermal rocket with an exhaust velocity of approximately 9 km/s, and so even assuming a "running start" air speed of 1 km/s, the mass ratio (mass ratio is the mass of the rocket with fuel divided by the mass of it with its tanks empty) it would need to achieve such an ascent is more than 20. This means that Jupiter is off limits for He3 mining, because it's probably impossible to build a hydrogen-fueled rocket with a mass ratio greater than 6 or 7. In contrast, with the help of lower gravity and still large equatorial rotational velocities, NIFTs with buildable mass ratios of about 4 would be able to achieve orbit around Saturn, Uranus, and Neptune.

CREATING A NEW MARS The control of nature that late twenty-first-century human civilization will possess will enable engineering works on a massive scale, with the most profound being the terraforming of Mars. The firststeps required in the terraforming of Mars can be accomplished with surprisingly modest means. The Martian soil contains enough carbon dioxide (CO 2) gas to give the planet an atmosphere with about a third the pressure of Earth's at sea level; if the planet were warmed slightly, this material would begin to outgas. Such a warming (about 6C) can be achieved by placing on the planet's surface factories that manufacture about 300 tonnes of supergreenhouse-effect halocarbon (CFC) gases per hour, an output rate that would demand a mere 1500 MW (0.0015 TW) of power. As the CO2 emerged from the soil, it would add its own potent greenhousing capability to the atmosphere, accelerating the warming process like an avalanche until temperatures in the Martian tropics during the

Colonizing the Outer Solar System

•

91

summer approached the freezing point of water. Such a Mars could allow some plant life to exist in the open in the tropical regions, and humans could walk abroad without spacesuits (needing only warm clothes and oxygen masks). However, the atmospheric levels of oxygen and nitrogen would be too low for many plants, and if left in this condition the planet would remain relatively dry, as the warmer temperatures took centuries to melt Mars' ice and deeply buried permafrost. It is during this second phase of terraforming Mars, in which the hydrosphere is activated, the atmosphere is made breathable for advanced plants and primitive animals, and the temperature is increased further, that human activity in the outer solar system assumes an important role. Activating the Martian hydrosphere in a timely fashion will require doing some violence to the planet, and one of the best ways chis can be done is with targeted asteroidal impacts. Why go to the outer solar system then? The reason, strange as it may seem, is that it is easier to move an asteroid from the outer solar system to Mars than it is co do so from the Main Belt or any other inner solar system orbit. This odd result follows from che laws of orbital mechanics, which cause an object farther away from the Sun to orbit it more slowly than one that is closer in. Because an object in the outer solar system moves slower, it requires a smaller velocity change (or LlV) to change its orbit from a circular to an elliptical shape. Furthermore, the orbit need not be so elliptical that it stretches from Mars to the outer solar system; it is sufficient to distort the object's orbit so that it intersects the path of a major planet, after which a gravity assist can do the rest. If these tricks are used, the required LlY needed to move an outer solar system asteroid onto a trajectory where it will hit Mars can be as little as 300 m/s. Moreover, because an outer solar system asteroid is likely to be made of ice, ammonia, and other frozen gases, these materials provide a ready-made source of propellant to move them. When heated to high-temperature vapor in a nuclear thermal rocket engine, such materials can generate gas streams with exhaust velocities of about 4000 m/s. This means that less than 10% of the asteroid's mass would be sacrificed as propellant in order to send the rest of it on its way. Consider an asteroid made of frozen ammonia with a mass of 10 billion tonnes orbiting the Sun at 12 AU, slightly beyond Saturn. Four 5000-MW nuclear thermal rocket engines could move it

92

•

Islandsin the Sky

300 m/s with 10 years of steady thrusting. This would be enough co cause it to swing by Saturn, after which it would coast for about 20 years until its impact. When the object finally hit Mars, the energy released would be about 10 TW-years, enough to melt 1 trillion tonnes of soil and release enough ammonia to raise the planet's temperature by about 3 C, incidentally forming a shield that would effectively mask the planet's surface from ultraviolet radiation at the same time. If one such mission were launched every year, in about half a century most of Mars would have a temperate climate, and enough water would have been melted to fill a sea 1 million km 2 in area 50 m deep. Only the potential energy of outer solar system objects can do such work. In comparison, the explosive power of thermonuclear weapons is puny; it would take 70,000 I-megaton hydrogen bombs to deliver the same punch as one of the small asteroids discussed previously.Even if so many thermonuclear explosives could be manufactured, their use would cause the planet to become unacceptably radioactive.

HIGH GROUND If swordscan be beaten into plowshares, the reverse is also true. The vastly superior firepower offered by the use of outer solar system objects over thermonuclear weapons will not escape the attention of military planners. It is a historic axiom of military science that he who controls the high ground controls the battlefield; in the relationship between the outer solar system and the inner planets, this axiom is doubly true since in this case the high ground comes fully equipped with an artillery battery that would outgun Zeus. It therefore followsthat not only economic, but military security from the mid-twenty-firstcentury onward will be impossible without control of the outer solar system. If the future should see humanity divided into various nations, then the outer solar system a century hence will see their bases, fortresses, and fleets. Warships of every major power will patrol the outer darkness to protect commerce and prevent any unauthorized attempts to move celestial objects. Addressing the needs of these outposts will create another form of interplanetary commerce, with high-technology products shipped

Colonizing the Outer Solar System

•

93

out from the inner solar system, and settlements of civilians springing up to produce the consumables and other goods that can be made lo, cally.

COLONIES IN THE COLD Just as in the old west towns spring up in the vicinity of mining camps, cavalry forts, and trading posts, so in the outer solar system true colonies will develop around the He3 mining operations, the military bases and their supporting shipping businesses. However, as elsewhere in the past, within a generation or two after their founding, such settlements will tend to take upon a life of their own. To those born on Titan or Miranda, the moons of Saturn or Uranus will be home, and few will have any greater desire to move to Earth than most terrestrials today have to live on Venus. It is not just a matter of sentiment; those raised in low,gravity environments, such as the moons of the outer planets, will be ill,adapted to live on Earth. Why be squashed to the ground on a crowded planet where even the native humans must travel by foot or vehicle, when you can live on the wide open spaces of Titan and soar like a bird? Furthermore, why live in a place where people must pay to have energy shipped to them from across the solar system, when you can live near the energy source and enjoy virtually unlimited power and all the advan, tages that come with it practically for free? Finally, why live on a planet whose social laws and possibilities are defined by generations long dead, when you can be a pioneer and help to shape a new world according to reason as you see it?

ROAD TO THE STARS No, humans will go to the outer solar system not merely to work, but to live, to love, to build, and to stay. But the irony of the life of pioneers is that if they are successful, they conquer the frontier which is their only true home, and a frontier conquered is a frontier destroyed. For the best of humanity then, the move must be ever outward. In the early twenty,first century, humans will settle Mars using ballistic transportation systems that will be, at best, marginal

94

•

Islands in the Sky

for the job. The growth of Martian settlements will force the development of fusion rocketry that will make the red planet commonplace, but give the pioneers the outer solar system in return. However, it won't end here; each step demands another. The farther we go, the farther we will become able to go. Ultimately, the outer solar system will simply be a way station toward the vaster universe beyond. For, although the stars may be distant, human creativity is infinite.

Chapter 6 Richard P. Terra

Islands in the Sky Human Exploration and Settlement of the Oort Cloud

After spreading thn 1ughl1ut the warm, compact inner core of the solar system, the human race will face the daunting challenge of crossing the deer ncean of space that lies between Sol and even the nearest ncighhoring stars. But just off shore, at the outer fringe of Sol's gravitational reach, countless small icy islands are scattered in a vast spherical archipelago reaching almost halfway to the Sun's nearest neighbors-the myriad comet nuclei of the Oort Cloud. Human exploration and settlement of the Oort Cloud will create a new environment for a diverse array of social, economic, and biological experimentation, and that diversity may indicate the future course of the evolution of our species. The Oort Cloud will also be the proving ground for the technology of interstellar travel: these tiny, scattered islands on the periphery of the solar system may be our first stepping stones to the stars. Although the notion of human settlement on comet nuclei dates back to the eighteenth century, it was not until the latter part of this century that the concept began to receive serious attention. In 1972, physicist Freeman Dyson offered this challenging vision: 95

96

•

Islands in the Sky

It is generally considered that beyond the Sun's family of planets there is absolute emptiness extending for light-years until you come to another star. In fact it is likely that the space around the Solar System is populated by huge numbers of comets, small worlds a few miles in diameter, rich in water and other chemicals essential to life . . . comets, not planets, are the major potential habitat of life in space. 1

More recently, astronomers, biologists, sociologists, historians, and design engineers have begun to subject the idea to rigorous-if somewhat speculative-scientific and engineering analysis. They have begun to explore the very real possibility of establishing human settlements in the distant reaches of the Oort Cloud, and what life might be like there. The Oort Cloud is thought to form a vast, spherical collection of small icy bodies orbiting at enormous distances from the Sun. Although direct observational evidence is sketchy, there are sound theoretical reasons for postulating its existence. It is generally believed that the comet nuclei in the Oort Cloud formed along with the inner solar system about 4.6 billion years ago. ln the most widely accepted theories, the solar system originated in the gravitational collapse of an immense cloud of interstellar gas and dust. As it collapsed the cloud flattened out into the classic solar nebula, a broad, fairly thin disk of material slowly rotating about the infant sun. Within the nebula, the gas and dust particles began to condense and clump together into small aggregates that grew slowly into larger and larger bodies, forming planets, moons, asteroids-and comets. Most comets are thought to have formed on the outer fringe of the solar nebula, at about the same distance from Sol as Uranus and Neptune (roughly 20 to 30 astronomical units [AU]). The bulk composition of these two planets is close to what wou Id be expected if a similar massof comet nuclei were collected together into a single, planet-sized body. In another line of evidence, most comets seem to be significantly depleted in carbon relative to the average cosmic abundance of that element. This can be interpreted as an indication that comets formed in a region of the nebula warm enough to evaporate much of the methane (CH4} present and prevent it from condensing with the remaining gasses and dust. As expected, comets are also badly de,

Islands in the Sky

•

97

pleted in hydrogen, as are all the other bodies in the solar system except for Jupiter, Saturn, and the Sun. Most of these small bodies would be swept up into the growing planets, but the orbits of many others would be perturbed and altered by close encounters. An encounter with Jupiter or Saturn would most likely result in complete ejection from the solar system into interstellar space. In the vicinity of less massive Uranus and Neptune, comet nuclei would also be perturbed into orbits far from the Sunbut not entirely beyond its gravitational reach. Initially, the comets would orbit more or less in the same plane as the inner system, but over time the faint tug from passing stars would randomize their orbital inclinations, creating a vast, dispersed, spherical cloud of small icy bodies-the Oort Cloud, named for Dutch astronomer Jan Oort, who first postulated its existence and structure in the 1950s. Estimates of the total number of comet nuclei in the Oort Cloud range from about 100 billion to several trillion-more than the total number of stars in the entire galaxy. Their total mass is difficult to estimate with any precision, but is probably at least equal to the mass of Earth, and may be much greater. It is thought that the original population of the Oort Cloud was about 2 trillion comets with a mass about 30 times that of Earth, but has since been depleted by escape into interstellar space and perturbation into orbits closer to the Sun, where the icy bodies evaporate after a time. The majority of the comet nuclei in the Oort Cloud are probably distributed throughout a broad spherical shell between 40,000 and 60,000 AU from the Sun (although some astronomers place the density peak a bit closer, at around 25,000 AU). This is just shy of one light-year (64,000 AU). It is possible that a substantial number orbit much closer to the Sun, between 100 and 10,000 AU, forming an inner Oort Cloud. Within about 10,000 AU of the Sun, the orbits of the comets would lie more or less in the plane of the ecliptic. Beyond that distance, the influence of passing stars is strong enough to produce the random spherical distribution of the outer Cloud. The outer periphery of the main Cloud is probably at about 100,000 AU-close to 2 light-years. The maximum dynamic limit is about 200,000 AU, where the gravitational pull of the galactic core begins to overcome that of the Sun (see Figure 6.1).

98

•

Islands in the Sky

1 li9ht Year(Ly) = 9.6 x 10 15 m (approximately6-4,000AU) 2 Ly

.·

.:-.

..

Figure 6.1. A schematic representation of the Oort Cloud.

The typical separation between comet nuclei in the Oort Cloud is probably about 20 AU: very widely spaced indeed. In the main Cloud the orbital periods would range from one to ten million years. Since the close encounters of Halley's Comet by Soviet and European spacecraft in 1986, a clearer picture of the nature of comet nuclei has begun to emerge. They are estimated to range in size from 0.1 to 10 km in diameter and in mass from 1013 to 10 19 g-between 10 million and 10 trillion metric tons. Though small in comparison to the planets, comet nuclei still contain huge amounts of material. Comets are thought to be made up of an extremely light, porous aggregation of the ices of various volatile elements and compounds, with a scattering of embedded dust and complex organic molecules. The ratio of dust to gas is estimated to range from 0. 7 to 0.85 by mass. The volatile component is thought to consist chiefly of water ice but also includes CO, CO2, methane (CH 4), ammonia (NH3), and a variety of other compounds containing oxygen, sulfur, and nitrogen. The ices are thought to include a high proportion of clathrates, in which other molecules are trapped inside the lattices of the frozen gases. The grains of dust and organic material may be remnants of the interstellar cloud or the condensing solar nebula. A breakdown of the estimated abundances of the chief elemental and chemical constituents of a typical comet is presented in Table 6.1.

•

Islands in the Sky

99

Table6.1. Composition of comet nuclei: estimated abundances and distributions

ElementalAbundancesa Element Number(%) H C N

Mass(%)

of Principal Mass & Volume Distributionb Components Component

Mass(%) Volume(%) 21

8.6

7.7

Silicates Carbon (graphite)

6

Very Complex

19

3.4 21.0

43.9 6.7

4.2

2.7

3.8

0

40.2

62.5

s

1.0

3.1

H2O

19

27.0

Mg

1.9

4.5

co

10

Si

4.9 8.8

Other

25

Fe

1.8 1.6

13.0 27.0

Ni +Cr

0.1

0.6

Al

Organics

Other Volatiles: CO 2, CH 4, N2, NHJ, HCN, etc.

0.4

• After Delsemme, in Wilkening (ed): Comets, p 122. ~ After Greenberg, in Wilkening, p 157.

The average density of cometary material is estimated to be about 1 to 1.5 g/cm 3-close to that of water (1.0 g/cm3 ). This material is thought to be a porous matrix of solid grains and larger particles, with the gaps between them filled with ice. Comet nuclei are likely to be irregular in shape, uncompacted and undifferentiated: For a body under about 10 km in diameter, the force of gravity is simply too weak to affect shape or structure to any significant degree. The aggregate would have a light, very fragile structure resembling that of windblown snow. Both structure and composition are expected to be fairly homogenous throughout, although the surface layers (to a depth of perhaps 10 m) may undergo some alteration due to sublimation of ices, irradiation by cosmic rays and other radiation, and occasional sandblasting from interstellar gas and dust. This light fragile structure means that the resources present in the comet nuclei will be readily accessible to any human settlers. The porous mixture of dust and ice would offer little mechanical resistance, and the two components could easily be separated by the application of heat. Volatiles could be further refined through fractional distillation while the dust, which has a high content of

100

•

Islands in the Sky

iron and other ferrous metals, could easily be manipulated with magnetic fields. Little additional crushing or other mechanical processing of the dust would be necessary, and its fine, loose-grained structure would make it ideal for subsequent chemical processing and refining. Comet nuclei thus represent a vast reservoir of easily accessible materials; water, carbon dioxide, ammonia, methane, and a variety of metals and complex organics. Although comets offer a huge reserve of easily accessible resources for any would-be settlers, they will be useless without a readily available supply of energy. Conditions at those distances from the Sun will be bleak: the Sun will not even be the brightest star in the deep black sky, and the temperature is a bare 10 K. Making the cold, barren ice islands of the comet nuclei habitable will be a formidable task, and energy consumption is likely to be very great. Yet energy will be the scarcest resource in the Oort Cloud. The size of any comet-based communities will be closely constrained by two factors: the availability of energy sources and the rate of energy consumption. Access to a dependable, long-term energy source will be crucial to the establishment of a viable, selfsufficientsettlement. Energy consumption levels will affect both the size and complexity of the economy, the standard of living, and longevity of the community as a whole. What are some of the potential energy sources likely to be available? Nuclear power-fission-is unlikely. Heavy elements such as uranium will be present in only miniscule amounts, certainly not enough to form the basis for a comfortable, long-term settlement. In addition, fissionprocessesare "dirty" in the sense that they generate a relatively high flux of radioactivity that may be undesirable. Another potential energy source is the controlled combustion of hydrogen and oxygen, which both occur in abundance in the vast amounts of water ice that makes up about one-fifth of each comet's mass. The catch, of course, is that water must first be hydrolyzed, an energy-consuming process. Settlements closer to the Sun, say within 1000 AU, or those with sufficient surplus energy from other sources, may employ Hz/02 reaction cells for some purposes. Hydrogen combustion will probably serve as the major propellant for longdistance travel, but is unsuitable for basic power generation. There is an alternate means of using hydrogen as an energy source-fusion. It seems reasonable to assume that the technology

Islands in the Sky

•

101

for fusion employing deuterium, a heavy isotope of hydrogen that fuses relatively easily, will be available by the time settlement of the Oort Cloud begins. Estimates of the ratio of deuterium to simple hydrogen in interstellar gases vary from about 1 in 10,000 to 1 in 100,000.Although rare, for an average,size comet this represents about 50,000 to 100,000 metric tons of deuterium. This enormous store represents enough energy to support a popu, when the supply is gone, lation of millions for several centuries-but those millions will perish. It seems much more likely that Oort Cloud communities will choose co support much smaller populations for much greater lengths of time, husbanding their precious supply of deuterium for perhaps thousands or tens of thousands of years. They may also employ a more efficient alternative fusion pathway involv, ing a mixture of deuterium, tritium (another hydrogen isotope), and helium to extend the lifetime of the available supply. At some point humans may also learn how to initiate proton fusion and thereby assure a virtually unlimited supply of energy employing simple hydrogen, the most common element in the cos, mos. But as yet, only the stars themselves have accomplished chis on a large scale. Finally, there is one alternative renewable source of energy avail, able: starlight. The concept of harvesting starlight to support comet, based communities was first put forward by astrophysicist Eric Jones and anthropologist Ben Finney 2.3: In principle all one needs is a giant mirror to concentrate the thin photon At half a parsec [about stream. The mirrors would be truly gigantic .... 1 1/2 light-years) from a Sun-like star the mirror would have to be about 1500 kilometers radius to collect a megawatt (roughly the size of the continental U.S.). Practicality and common sense might well dictate a larger number of much smaller collectors.

Jones and Finney estimate a typical medium,sized comet could supply enough aluminum for mirror surfaces to collect about 300 megawatts (MW) from starlight. If additional metals were used, this capacity could be enhanced somewhat. Jones and Finney estimated per capita power consumption at 0.6 MW, about 30 times the current U.S. rate of 20 to 25 KW per capita; thus the typical population of a comet,based community would be about 500.

102

•

Islands in the Sky

The amount of aluminum and other reflective metals in a given comet will set a definite upper limit on the area of mirror surface that can be created, and thus on the population that can be supported by collecting starlight. Comet-based communities closer to the Sun in the inner Oort Cloud will therefore enjoy a higher photon flux and will have more energy available to them, and will be able to support larger populations. Although the energy supply will be the most important limiting factor for any comet-based settlements, there are a host of other factors that will influence the nature of their economies. Most of these communities are likely to be small and isolated, with populations of only a few hundred. Such small economic systems are sure to face many limitations and constraints. Many types of tools, specialized pieces of equipment and expert skilled labor will all be in short supply. The economy will essentially be a closed system, especially for the more distant and isolated settlements. For many items and commodities, supply will not be related in any way to need or demand. The members of these communities will of necessity be very dependent upon one another and are also likely to develop a strong conservation ethic; they will recycle as much as possible. Their relative isolation will encourage self-reliance and cooperation, as well as the pooling of skills, labor, and resources. Small, isolated, tightly closed economies are thus likely to be more communal in their outlook, particularly in regard to resources, power consumption, and the allocation of labor. Larger communities will probably have a wider variety of economic patterns. Because populations are likely to be small, each individual will be of immense value to the community, representing a pool of knowledge, skills, and experiences, as well as a heavy investment in training and education. It will be necessary for the settlers to become generalists: A small community will simply not have enough people to support a specializedexpert in every field. Each person will have to develop a sound working knowledge in many overlapping fields, and consequently will have less time to devote to becoming expert in any one of them. Some specialties may have to be eliminated altogether in very small communities. Many of these limitations and constraints can likely be alleviated or even eliminated with the aid of sophisticated technology.

Islands in the Sky •

103

The exploration and settlement of the Oort Cloud, the establishment of viable communities in the cold deeps of space, will demand heavy reliance on many different forms of high-technology equipment. A combination of the right tools and skills to use them will be vital. It seems unlikely that a small band of pioneers will be able to create a secure, comfortable niche in the Oort Cloud without the use of automated systems and artificial intelligence (Al). Such systems will eliminate the need for human intervention at many basic but routine levels of activity, freeing people to apply their skills where they are needed most. Computers and automated systems are skilled tools which can enhance and multiply human capabilities manyfold. The human residents of the Oort Cloud will control and monitor the operation of a complex array of self-directed systems. Most of the basic comet-mining and materials-processing operations will probably be automated. They will produce stockpiles of refined materials and, at the highest level of complexity, produce many types of finished products. Artificially intelligent systems will also be essential to the viability of the communities, particularly for small settlements. They will rely heavily on expert systems-sophisticated computer programs that combine an extensive store of information with a powerful decision-making capability for interpreting complex data. They will operate at the level of a human expert that perhaps the community cannot support. The human members of the community, with their generalized skills and experiences, will be capable of handling all the normal operations and events; the expert systems would serve to supplement and complete the basic suite of skills and abilities required to keep the community healthy. Because of their dependence on Al systems, residents of the Oort communities will also need to be skilled in the manipulation and application of information. They will have access to the vast sum of human knowledge, but it will be useless to them without sophisticated information processing systems. They will have to become adept at working in cooperation with their Al systems to manage the masses of data available, to sift through it to extract what they need. The use of sophisticated computer-based systems will be a

l 04

•

Islands in the Sky

basic skill in every Oort Cloud community, and it seems likely that education will be based in large part on AI systems. All these factors will have a number of important effects on the shape of the settlement's economy. The distribution of activity will be very different for any Earth-bound economic system. The primary sector of the economy-the exploitation of natural resources-is likely to be small and almost completely automated. Human involvement will be minimal. The primary sector will consist of two basic activities: energy production and the harvesting of cometary resources. Once the appropriate systems are established, both will be relatively simple activities. The secondary sector-the transformation of the natural resources-will include refining and processing the raw cometary feedstock, manufacturing, construction and assembly operations, agriculture and food production, and recycling. Again, many of these activities will be highly automated, but closer human supervision will be necessary to tailor these activities to the current needs of the community. It is the tertiary sector of the economy, support services, that will engage the majority of human skill and attention. It will include operation of the utility and life support systems, maintenance and repair operations, transportation, health care, information processing, storage and retrieval, education and administration, business and commerce. The settlers will have to choose between many different possible paths for development. Will they allow the community to grow, or should they impose controls on themselves to maintain a small population? Will they want their community to last many years at a low rate of consumption, or will they want to trade communal longevity for a higher rate of resource consumption? The outcome of these decisions will depend primarily on two factors: the available supply of resources and the standard of living that the community desires and will defend. The average standard of living is likely to be quite comfortable, within the limits of the communities' capabilities. Because of its extreme fragility, the surface of the comet itself will be unsuitable for permanent dwellings; the habitats will probably be free-flying. Physically, they may resemble the Bernal Spheres and O'Neill Cylin-

Islands in the Sky •

105

ders already envisioned for space-based communities in the inner solar system. The most significant difference will be the lack of large window surfaces, since there will be no need to admit sunlight for illumination, which will allow for an increase in the usable surface area inside the habitats. Communities lacking a large supply of deuterium or those choosing to conserve their supply would locate their habitats at the center of huge arrays of mirrors for the collection of starlight. A variety of social and economic factors suggest the members of such a community, which would number about 500, would be dispersed throughout the array in bands of about 25 persons-a dozen adult men and women and their children. Each band would tend a mirror farm stretching across perhaps 30,000 km of space, overseeing a complex network of robots and other automated systems responsible for the routine operation and maintenance of the mirror farm. In the event of an emergency requiring human intervention, the settlers would need to be able to reach even the edge of the array quickly, and so will gather the mirrors in tightly packed arrangements. The individual mirror farms of the separate bands would then be clustered into a vast array supporting the community's entire population, spread out over 150,000 to 200,000 km of space: an immense surface area of ultra-thin metal film centered around the invisible point of the comet nucleus, gathering the faint light of the stars (see Figure 6.2). Communication between bands within the community could easily be maintained by radio and laser. Despite the limitations of the environment, Oort Cloud communities that choose to grow and develop large economies will have many options. One of the most obvious ways to enhance their resource base will be to create clusters of comet nuclei. Clustering would require the commitment of significant amounts of time, energy and resources, but the potential rewards will ensure that some communities will begin to move additional comets into more convenient orbits. Given the average separations and relative orbital velocities of comet nuclei in the Cloud, it would take between 5 and 10 years to bring a new comet into a cluster. The energy required represents only about 3 to 5% of the available deuterium supply; clustering

106

•

Islands in the Sky

"Tribal" errey ot 19-slngle bend mirror terms.

0--00

0000 " ,,, ____ •~oooo "gogo~ /0° 000 oo 0o 0oo // 000 oOoOc) o0 o '/ / ,I

1

0

1---30,000 Anll'f

83000km km-----,

-----200,000

km-------1

ot 1 MW mirrors to support band of 25.

Eachcircle represents en array or smaller mirrors. The habitat would be located at the center of the erray.

Figure 6.2. Schematic representation of "mirror farm" arrays for harvesting starlight. ( Illustration after Ref. 3.)

will be relatively economical once the necessary resources are committed to the project. Clustering will probably be common, if only because large comets will be rare. Clustering would allow relatively large populations to exist, especially in the inner Oort Cloud, where the photon flux from the Sun will be fairly high and consequent energy production levels will be capable of supporting much larger communities. Jones and Finney have estimated that the largest practical population for a cluster will be about 100,000, when the energy consumed in transport between the dispersed habitats of the cluster begins to strain the reserves of the community. Social factors may also limit the size of large clusterbased communities. In addition to clustering, there will be other means by which the settlements can enhance their economies. Their isolation in the depths of space will be a relative thing; they will be in constant communication with neighboring communities. The average separation between comets in the Cloud is thought to be a few light-hours; between settlements it will be perhaps a light-day. News, messages,

Islands in the Sky •

107

and other information will flow back and forth across the empty space between them. Some settlements, particularly the larger ones, may engage in some sort of trade, both with the inner system and among themselves. Initially, many communities will need to import those items that they cannot make for themselves. As they develop, they may consider more complex forms of trade; it is certain the decision to expend the time, energy, and resources in long-distance commerce will be a complex cost/benefit calculation. Items for trade might include specialized technology, bioengineered species specifically tailored for conditions in the Oort Cloud, computer and Al programs, and information itself. Perhaps some communities will specialize within a larger dispersed cluster or bloc of trading communities. Because of the constraints and limitation imposed by the environment, one primary social concern of the Oort Cloud communities will be the careful regulation of the health and size of their populations. Until the settlement has developed the skills and resource base to support growth, some form of control will be necessary to limit the size of the population. In very small, isolated populations, even the timing of reproduction may need to be regulated, in order to maintain an appropriate distribution of age and sex. Unless communities adopt fairly stringent measures, the population will inevitably grow. The members of the settlement will then have to choose between a higher rate of resource consumption, making a commitment to enhancing their resource base, or assisting their excess population in establishing daughter settlements around other comet nuclei. Both static and growing communities will also need to be concerned with the genetic health of their populations. Within the small, relatively isolated gene pools of the Oort Cloud settlements, population genetics will become a factor that will affect both everyday life and the long-term viability of the community. Very early on, when a new community is being planned or established, a major concern will be the "founder effect"-the skewing of the genetic composition of a population because its founding members are very few. The danger exists that the founding members of a community may share too high a proportion of deleterious genes that will soon become reinforced in subsequent interbreeding. This

108

•

Islands in the Sky

problem could be largely eliminated by careful screening during che planning and recruitment stages for a new settlement. New colonists would in part be chosen to maximize genetic variation in the new population, providing a sound base for healthy future growth. Founding populations can be very low-perhaps as low as 10 or 20 individuals-if the original breeders are genetically diverse and share few common deleterious genes. The negative effects of early inbreeding can also be avoided by social customs or ocher regulations that prevent bad crosses, particularly among close relatives. The potential negative effects of inbreeding will remain, however, as the settlement's population grows and becomes more uniform, but will be of major concern only to very small, isolated groups. Boch natural and artificial selection will come into play before inbreeding could cause serious damage to the health of the community. In nature, inbreeding often results in a high level of spontaneous abortions and early deaths; natural selection operates co remove the bad crosses from the breeding population. In healthy communities, matings would be structured either by intent or by social custom to minimize the negative effects of inbreeding. It is also important to remember that genetic uniformity need not be a problem provided chat any deleterious genes are eliminated from the founding population and that any new negative traits arising by mutation are culled or otherwise prevented from breeding. Genetic engineering techniques may also be employed to maintain a healthy gene pool. The human consequences of such requirements for the maintenance of the community's genetic health will be great. The social tensions generated by the practice of eugenics and genetic engineering may also be difficult to manage. However, there are other options for communities that wish to avoid negative inbreeding, including outbreeding with other settlements. Outbreeding has the obvious positive effect of increasing genetic variability. But its practice also serves as a mechanism for complex social interaction as well. Prior to the terrestrial agricultural revolution, people tended to live within small hunting and gathering bands of about 20 to 30 men, women, and children. It has been noted by various social anthropologists that groups of this size still seem to be the most congenial to humans worldwide in both their private and professional

Islands in the Sky

•

109

lives. This preference is no doubt part of our physical and social evolutionary heritage. But humans also tended to form larger social units, gathering into clans or tribes numbering about 500 or so. Normally dispersed as individual bands, the tribes maintained social cohesion through intermarriage. Indeed, the size of such groups is just about the mini, mum number needed to maintain a healthy diversity within the common gene pool and to avoid inbreeding. It is possible to anticipate, then, that the dispersed communities of the Oort Cloud may form similar social links. A single community tending an extensive mirror farm as described earlier could form a viable, self-contained breeding population by marriage exchanges between bands. In addition to preserving the genetic health of the community, such exchanges would serve to link the bands into a large, cohesive social unit. Just as dispersed tribes on Earth gathered periodically for important social and religious observances, so too might the dispersed bands of a comet,based settlement gather their entire population on occasion. Outbreeding would also allow individuals whose reproduction might otherwise be restricted or forbidden to find an alternative, providing an outlet for any tensions that might arise. And if such a place could not be found within the tribe, there is also the possibility of exchanges with other communities around other nearby comets. Such exchanges would be more expensive in terms of time and energy, but would serve to cement social ties between settlements. The necessity for monitoring gene flow within a given population, along with all the other shaping influences of the deep space environ, ment, will have profound social consequences for the Oort Cloud communities. Habitat residents will be able to set temperature, humidity, day length, the weather, atmospheric pressure, even the level of gravity according to their collective preference and the limitations of human physiology. They may even choose to live in zero gravity. Lifestyle and social values, the entire "world view" of the Oort Cloud settlers will be very different from those of any terrestrial culture. Without diurnal or seasonal variations, their sense of time will differ. Their perceptions and use of space may also differ, and the ways in which private and public volumes are used and arranged thereby altered. In a tightly closed artificial environment, attitudes

llO

•

Islands in the Sky

toward clothing and nudity are likely to change radically, as is the concept of privacy. In personal relationships, love, sex, and mating will all be quite separate. Family structures are likely to be very different when marriage or cohabitation preferences may differ from permissible genetic crosses. Breeding will be much more of a social decision than it is on Earth. Given a basic band-and-tribe social structure, extended families are likely to be common. Communal or collective family arrangements may also be common. Social, political, and administrative systems will also change, and new ones will arise. They will be shaped not only by the ageold needs and limitations of human social interaction, but also by wholly new factors unique to the deep space environment. Settlers in the Oort Cloud will face the exciting challenge of creating workable societies from whole cloth. As always, the basic social conflict within these communities will be between the needs and constraints of society as a whole and the personal freedom of individuals. Members of these small, isolated communities will be faced with a variety of constraints. Restrictions on breeding practices may be imposed by the community in order to maintain a balanced, genetically healthy population. In very small settlements, even the timing of births may be regulated in order to preserve a balanced distribution of male and female, young, middle-aged, and elderly. Communities aiming for a low or static population growth rate may impose strict absolute limits on breeding until the resource base has been developed to support larger numbers. Similarly, young people born and raised in an Oort Cloud settlement may face fairly tight constraints in their choice of career or vocation. As in all economies, the needs or demands of the community will determine the types of work available; career choices may often be a matter of timing and happenstance. In small communities, where age and gender distributions may have to be carefully controlled, career choices will be limited for the limited number of youths available for apprenticeship: They will be pressured to take the positions that need to be filled as they come of age. New settlements with undeveloped capabilities, and smaller settlements with few resources or a limited energy supply may also impose some restraints; small, isolated communities will probably

Islands in the Sky

•

111

be conservative in approving unnecessary expenditures of scarce resources, thus limiting the range of possible private and profes, sional activity. Social values are likely to have an enormous shaping influence on the types of social systems that develop in the distant Oort Cloud communities. These settlements, perhaps comprised of highly interrelated clans, are likely to place a high value on cooperation, consensus decision making, and compromise. In the hostile and unrelenting deep space environment, where cooperation and mutual interdependence will be crucial to survival, they will feel a strong need to maintain social harmony. Members of such tightly knit communities will face a variety of social pressures: They will be part of a symbiotic network of duties and obligations. There will be ties to family and to tribe or clan, in addition to the community as a whole. Complex interrelationships will arise when emotional and sexual relations must be separated from breeding, as well as when family and friends emigrate to other habitats and communities. As in Earthbound societies, education, in addition to training each person for the complex task of deep space survival, will probably also serve as a primary means of inculcating the community's social values. But there are other factors to counterbalance these pressures and constraints, to allow for the preservation of individual freedoms. In a community where every individual represents a heavy investment in education and training, and a valuable pool of badly needed skills and experience, rigid social controls such as ostracism, imprisonment, or capital punishment seem unlikely; a large range of social behavior will have to be tolerated. Coercion would be difficult at best, espe, cially in very small communities with few resources or personnel to spare for "police actions"; nonconformists and dissidents must be allowed to find or to create an acceptable niche within the commu, nity. Larger settlements, in which any one individual plays a less vital role, may develop more rigid social structures. The Oort Cloud communities will need to develop social and political mechanisms to ensure the freedom to express diversity and individual choice, to avoid oppressive measures aimed at enforcing unnecessary conformity to arbitrary community standards, and to eliminate or at least alleviate the negative consequences of commu, nity decisions.

112

•

Islands in the Sky

It is likely that a wide variety of social systems will develop among the Oort Cloud settlements. In some, social pressures may win out, and a rigid, authoritarian social structure created. But the need for consensus and close cooperation among community residents may help prevent the development of oppressive, highly stratified societies, at least among the smaller settlements. Similarly, the necessities of cooperation and mutual dependence will temper any tendency toward extreme individualism. Individuals will be expected to take responsibility for their actions and their role in the community; both wild individualism and blind obedience to social convention would be frowned upon. In such a social context, within a culture that values diversity, mutualism, and social harmony, current Western-style democracy may be considered inappropriate. Despite constitutional guarantees, majority rule often enforces homogeneity and suppresses minority viewpoints. Democratic systems, in which individuals are expected to "vote in their own interests," might also foster a divisiveness that small, isolated communities could ill afford. The smaller band-and-tribe settlement may favor consensus deci• sion making, in which minority and dissident viewpoints can be more easily accommodated. But such systems also have a drawback: Becauseeveryone must agree to the joint decision to reach a consen• sus, community members may feel pressured to "go along with the group." If the pressures became strong enough, they might foster a dangerous and oppressive homogeneity. Another communal decision-making system that often appears among communities of closely-related, tribal people is the "hardship· on-no-one" arrangement. In such a system, the emphasis is not so much on the final decision reached as on the consequences of each option. Decisions are often based on finding the option that places the least burden on the least number of individuals, and working to alleviate even those negative results of communal decisions. For example, youths guided toward a career specialty the community deems important might be allowed to pursue secondary specialties of their own choosing. A drawback to this system as well as for consensus decisionmaking systems is that they are often very time-consuming processes. As on Earth, the Oort Cloud communities may choose to deal with some social pressuresby displacing them spatially. Nonconform-

Islands in the Sky

•

113

isrs and dissidents in one band may find a comfortable and rewarding social niche in another habitat within their native tribe. Although a single clustered community would have a pool of shared communal values, each habitat within the settlement might have its own indi, vidual system, offering enough diversity to accommodate and allevi, ace most social frictions. As a last resort, migration between settle, ments would provide a social safety valve. le thus seems likely chat a variety of social systems would arise among the Cloud settlements. They will have co respond to numerous needs and demands, seeking to find a balance between maintaining a stable, functional society and personal freedom for individuals. The size of the community will be an important factor in its social development. The larger Oort Cloud settlements will probably offer a wider range of social forms. They may be able to accommodate more competition and divisiveness. Large communities, however, are also difficult to administer by consensus, and may employ a form of electronic democracy. Clustering, trade, and the development of the settlement's resource base and capabilities would create opportu, nities for the creation of more diverse and complex societies. A mosaic of many different social forms, distributed throughout the settlement, seems possible. But even the largest of the Oort Cloud communities will be quite small in comparison to the societies of billions that will develop in the inner solar system. As the relatively compact inner core of the system is developed, it will become more tightly integrated socially and economically, more centralized, and its culture more homogeneous. Although it may cake some centuries, it is likely chat mass cultures with populations of billions will eventually develop. The small, dispersed communities of the Oort Cloud will remain largely independent of the inner core government and societies. Their potential for diversity may offer a refuge for dissident groups seeking to leave the mass societies of the inner solar system. Just as the relative isolation of the Oort Cloud settlements from the inner system and from each ocher will foster social divergence and cultural diversity, so, too, will it create the conditions necessary for biological divergence and evolutionary change. As we know from terrestrial experience, reproductive isolation fosters such changeas Charles Darwin observed when he visited the Galapagos Islands

114

•

Islands in the Sky

as a young man. Diversity also fosters adaptive radiation-the open, ing and exploitation of new niches in the environment. Many factors are likely to encourage human evolutionary diver, sity in the Oort Cloud. Despite all efforts, the founder effect and random drift will cause some changes in the small isolated gene pools. Fortuitous beneficial mutations may be retained and encouraged to spread, and perhaps enhanced by artificial means. Eugenics and ge, netic engineering are likely to be well integrated into the social and economic life of the settlements, both as a means of maintaining a healthy gene pool and in the production of goods and trade items. Bioengineered products may be common. Social values will also be very different, and the distant Oort Cloud communities will feel much less keenly the social, cultural and governmental pressures that might prevent biological experi, mentation among the inner system cultures. Their environment and lives will be very different; they may choose to remake themselves in their own image, to adapt themselves to the needs of living in deep space. Estimates of the time required for natural selection to produce a new species of human being range upward toward a million years. However, in the small, isolated gene pools of the Oort Cloud settlements, perhaps encouraged by directed breeding practices and human genetic engineering, it would not take long for any chances to become widespread. Minor changes might take a few generations-perhaps a century-to become entrenched. Major structural or biochemical alterations would probably be attempted more slowly, on a scale of many hundreds or thousands of years. But it is clear that a new species of genus Homo could appear within about 10,000 years. The exploration and settlement of the far reaches of the Oort Cloud will also have a profound effect on the course of human migration to other star systems. Simply reaching the Oort Cloud and establishing viable, long-term settlements there will spur the development of the technology required for interstellar flight. Al, though the propulsion systems needed for journeys of tens of hundreds of light-years will differ from those used to travel only 1 or 2, the opening of the Oort Cloud will serve as the proving ground for such systems, as well as the development of closed artificial ecosystems. The social experiments of the Cloud communities may also play an important role in starship design.

Islands in the Sky

•

115

Though many interstellar missions may be launched by the large and wealthy inner solar system cultures, others may depart from construction sites or launch points in the far reaches of Sol's Oort Cloud, which stretches out to perhaps 2 light-years-a significant head start. Such ships may travel directly across the deep oceans of space to reach other stars, but others may visit way stations during their journey: comets drifting in interstellar space.4 Estimates of the total amount of material ejected from the solar system during its formation and subsequent evolution range from as low as 5 to 25 times the mass of the Earth up to 100 or even 1000 Earth masses. Presumably, other stars would produce roughly similar numbers of rogue comets. Interstellar space should be strewn with small, isolated islands. Distances between these drifting islands are difficult to estimate. They may be no more widely spaced than the comets of the Oort 20 AU. Other estimates range up to hundreds of Cloud-about thousands of AU between these wandering interstellar comets. Some of these interlopers are bound to be found within Sol's own Oort Cloud, working their way free of the Sun's grip, at perhaps one in every several thousand. Relative velocities between solar and interstellar comets will be low, and it will be tempting for Oort farther out into Cloud resiJents to hitch a ride outward-perhaps the permanent Cloud, perhaps out into interstellar space. Drifting outward at about 10 km/s or about 2 AU per year, they will make slow progress indeed. In 50,000 years they will be halfway to the nearest stars. But by then they will be wholly adapted to life in interstellar space and will perhaps not be too concerned with visiting other star systems. One could imagine the human species slowly spreading throughout the galaxy in this way, but it seems more likely humanity will move from star to star by fast ship, followed by slowly expanding waves of comet-based settlements meeting and overlapping in the voids between the settled scars. Ocher, more daring Oort Cloud communities may choose to take their home comet with them. If the slow outward drift seems too slow to them, and fast ships too expensive, they may attempt to launch themselves out of the solar system via a gravity-assist maneuver close to the Sun. 5 Large rocket engines burning hydrogen and oxygen, or perhaps a mass driver, could be used to dissipate the comet's orbital velocity

116

•

Islands in the Sky

using only a fraction of the comet's resources. During the long fall toward the Sun, the adventurers could use the rapidly increasing flux of solar energy to produce huge stores of hydrogen, to be consumed in fuel cells to provide power during the long interstellar cruise. The mirror farms or other solar power collectors would also serve to shield the comet nucleus from the Sun. After curving around Sol, the comet would leave the system with a velocity of 100 to 150 km/s, or about 1 light-year every 2000 years. Travel times between star systems would be on the order of 10,000 years-far shorter than that of drifting "wild" comets. During the interstellar journey the settlers would live as they had in the Oort Cloud, relying on their comet's stock of deuterium or on hydrogen produced when close to the sun. In the target system, a small band of colonists would be dropped off aboard habitats constructed during the long cruise, before rounding the new star and heading off toward a new destination. It thus seems likely that human expansion into the comet archipelago of the distant reaches of the Oort Cloud will only be the first step toward expansion to other star systems. Comet-based settlements will be humanity's stepping stones to the stars, across the far-flung islands in the deep ocean of interstellar space.

Chapter 7 Dr. Robert L. Forward

Alien Life Between Here and the Stars

Life is ubiquitnus. We find it in every nook and cranny Earth possesses. One goal of space exploration is co find life on some body ocher than Earth. Using our robot proxies, we have explored some of the more likely rlaces for life in the solar system, but presently, it appears as though there is no alien life in the solar system. If this is found to be true, then the usual conclusion is that the nearest alien life will be found on an Earch,like planet around some distant star. It is possible, however, that alien life exists between here and the stars. Beyond the planets lie the comets. Some of them are so large that even though they have ice,covered surfaces like comets, they have rocky cores like asteroids. They can be considered "comet, oids"-half comet and half asteroid. In 1992, a large cometoid was found in the Kuiper Belt with an orbit at 42 AU, far beyond Neptune, which orbits at 30 AU. Since then many more cometoids have been observed. The Kuiper Comet Belt merges into the Oort Comet Cloud, a spherical collection of ice,covered bodies loosely bound to the Sun, extending to the nearest stars. In the previous chapter on human exploration and settlement of the Oort Cloud, Richard P. Terra discusses the difficulty of living 117

118

•

Islands in the Sky

in a region of space far from any source of light. What could be a viable source of energy out in those cold depths? One answer is to convert the metals in a cometoid into large thin-film mirrors to collect starlight. Terra doesn't believe that nuclear fission energy is viable, but the most common nuclear fuel, uranium 235, is relatively abundant and is found in nearly all rocks. In comet ice, where most of the volatiles have evaporated away, the mass abundance of uranium 235 is about one part per billion. That means, in a comet 1000 meters in diameter, with a mass of a billion tons, there is a ton of uranium 235! There are other sources of energy in comet ice. The ice is continually bombarded by high speed cosmic rays that create ionized atoms known as free radicals. On Earth, where the average temperature is near 290°above absolute zero, these free radicals quickly recombine, releasing their energy as heat. In the Oort Cloud, where the average temperature is 20°above absolute zero, the free radicals remain frozen in the ice, ready to release their energy. Also, the grains of dirt embedded in the ice contain many radioisotopes that emit charged particles called alpha and beta particles. These also create free radicals within the ice. The problem with these Oort Cloud energy sources is that, like starlight, the energy is spread out over a large volume, making it difficult to collect. The method for collecting starlight is obvious-produce large mirrors. The methods for collecting uranium 235, radioisotopes, and free radicals are not as obvious. It is unrealistic to expect human beings to power an Oort Cloud civilization by collecting these alternate sources of energy, but the right species of alien could thrive and reproduce on these energy sources. For my novel1 Camelot 30K, I created an alien species called the keracks-intelligent, civilized creatures the size and shape of a crawfish-living on a cometoid they call lee (of course!). Because the cometoid has a near vacuum for its atmosphere and a surface temperature of 30 K, or 30° above absolute zero, I raised the internal pressure of the keracks to five atmospheres and their body temperature to 70K. To hold the pressure differential, I designed them with strong boron carbide carapaces. For their blood I chose F2O-difluorine oxide-a yellow-brown liquid that melts at 49 K and boils at 128 K, so it is liquid at kerack body temperature. This fluid is explosive when heated, which is put to good use in the story.

Alien

Life

Between

Herc

and

the

Stars

•

119

Next, I needed a scientifically plausible method of extracting and concentrating the dilute energy sources. I designed the kerack "cities" along the lines of an ant colony. Like ants, the keracks are merely specialized "units" of the "city organism." The keracks are the motiles that direct and tend the other units, and also collect and shape the materials that are gathered into the city. There are other specialized units; including a fungus that spreads through the ice, obtaining energy from the free radicals and extracting and con, centrating such nutrients as radioisotopes and metals. There are iceworms that eat this fungus, further concentrating the nutrients anJ elements, which they can separate by isotopic mass. These ice, worms are the primary "meat" source for the keracks. There are multiberry plants which concentrate metals and inorganic compounds, again separated into different isotopes. There are large beasts of burden, called heullers, that perform the major excavations and also supply meat. Then there is the Queen, whose task it is to lay eggs. The alien equivalent of DNA in these eggs contains the prescription for each different unit. What the larvae become when they hatch depends on what they are fed. Almost all the heavy elements that pass through the "belly-button" mouths of the keracks is gathered into a metal ball are excreted, except for one-which in their midsection, called the "soul." The soul of each kerack consists of pure uranium 235. The major energy source of each "city" is the slow nuclear fission of the uranium in each kerack soul. The "city" is a living nuclear reactor! The boron in the kerack shells is pure boron 10. When boron 10 captures a neutron, it fissions into an alpha particle and a lithium, 7 nucleus, releasing energy. Thus, it converts nuclear energy in the form of free neutrons into heat to keep the kerack body warm. At the time of "fruition" of the city organism, the keracks produce the gamete units needed for reproduction of the species, sexually cross-combine them, and transform them into spores, which are then scattered into space so that the species can find new cometoids on which to grow. For small cometoids, the high pressure inside the kerack bodies squirts the spores to escape velocity. For larger comet, oids, the explosive power of the kerack blood drives the spores into space. For extremely large cometoids, an even more powerful explosive force is available. . . .

120

•

Islands in the Sky

If I, a mere human, can, in a few months, design an alien life form that could live and reproduce in the Kuiper Comet Belt, then Nature, given billions of years, can certainly create even better examples. Hence, when we venture out to explore the comets, we should be prepared to be surprised. The comets lie at one end of the mass spectrum. Astronomers now have a good idea of the mass distribution of objects in deep space. There are many small objects such as comets and a few large objects such as galactic black holes. From these statistics, astronomers estimate that there are moderate numbers of medium,sized objects like cometoids, planetoids, gas giants, and those unlit subscars called "brown dwarfs." A brown dwarf is a massive body that is bigger than Jupiter and smaller than a star. It has been estimated 2 chat there should be dozens of brown dwarfs between here and the stars. All are potential targets for precursor interstellar probes, for there may be life forms on them. The outer cloud layer of a brown dwarf is cold; the central core is hoc. Somewhere between the cold surface and the hot center lies a region where water is a liquid. On Jupiter and Saturn, chis region is down in the third cloud deck, at a pressure of 5 to 10 arm, where the clouds are composed of water drops. Carl Sagan has written many papers 3 in which he not only describes, but also actually "designs" various life forms that could reproduce in the darkness under the cloud decks of the gas giant planets in the solar system. He has designed "sinkers" that are carried upward by convection currents when they are young and small in diameter, but then begin to sink as they mature and grow larger. As long as they have time to reproduce new small young before they fall down to depths where the temperatures are so hot that they are "pyrolized" (fried to a carbon crisp), then they are a viable life form. Sagan also describes "floaters" that are large enough to use hydrogen,filled gas bladders that allow them to float in the slightly more dense atmosphere of 94% hydrogen and 6% helium. The float, ers feed off the sinkers. Sagan also describes "hunters" chat feed off of both the floaters and sinkers. The hunters could be huge "air whales" the size of an Earth whale, perhaps even as large as a city or state. (This sounds interesting; I think I'll write a book about an intelligent air whale on Saturn that is as big as Central_ Park.) As

Alien Life Between Here and the Stars

•

121

the source of energy for his air creatures, Sagan mostly assumed that some photocatalytic process was used, either chemical photosynthesis inside a sinker or photoproduction of energetic molecules within the atmosphere. He also designed life forms that don't need sunlight and which could live off of energetic molecules welling up from the depths of the planet. Ethane, for example, has an excess of free energy compared to methane, which is the most common carbon molecule at equilibrium within the atmosphere. The giant red spot on Jupiter is evidence that there are giant upwellings of exotically colored unknown compounds on large gas giants. We know of life forms here on Earth that don't require sunlight to thrive and reproduce. These are found in the hydrothermal volca, nic vent fields along the Mid,Ocean Ridge. In those vent fields are mineral,rich springs of super,hot water supporting cases of life: long tube worms, miniature crabs, and blizzards of bacteria, all surviving, not on sunlight, but the chemical energy in the hot minerals and energetic chemical compounds, such as hydrogen sulfide, gushing from rhe vent springs. The tube worms are the most alien lifeform that dwell near the hydrothermal vents under Earth's oceans. They have no mouths or digestive tracts. Instead, they have a symbiotic relationship with sulfur,eating bacteria living inside a sac within the interior of their bodies. These worms have gill,like plumes on the portion of their bodies chat extends from their protective shells. Specialized blood in these plumes extracts hydrogen sulfide and carbon dioxide from the spring water, and carries chem co the sac containing the symbiotic bacteria. The bacteria oxidize the hydrogen sulfide with the carbon dioxide to produce organic carbon, which the tube worm then uses to build the hydrocarbon molecules it needs. If these alien creatures can exist and grow in the dark hydrother, mal vent fields in the ocean of Earth, then similar alien creatures could exist and grow in the dark hydrothermal vent fields in oceans on brown planetoids between here and the stars. We know the stars exist-we can see them. They are calling to us. But they are far, far away, and the journey to the planetoids they harbor and nurture with their light is a long one. One day we will go there and search for alien life. We do not know if there are suitable abodes for alien life between here and the stars, since we haven't been able to see any planetoids

122

•

Islands in the Sky

or brown dwarf substars yet. But, as we take our first baby steps on that long journey to Alpha Centauri, perhaps we can stop along the way at one of these potential abodes for alien life, anJ paddle around in the clouds and oceans to see if we can scare up something interesting to talk to.

Part Three

Creating New Worlds

Chapter 8 Robert M. Zuhrin ancl Christopher P. McKay

T erraf orming Mars

Not so long ago, the subject of human voyages to the Moon was the domain of science fiction. Today, lunar expeditions are a subject for historians, and manned Mars exploration the province of hardheaded engineers. A great many people now accept the idea that, in the not too distant future, humanity will have a permanently staffed base on Mars, or even a number of large settlements. However, the prospect of drastically changing the planet's temperature and atmosphere toward more Earth-like conditions, in other words "terraforming" Mars, still seems to the large majority to be either sheer fantasy or at best a technological challenge for the far distant future. But is this pessimistic point of view correct? Despite the fact that Mars today is a cold, dry, and probably lifeless planet, it has all the elements required to support life: water, carbon and oxygen (as carbon dioxide), and nitrogen. The physical aspects of Mars-its gravity, rotation rate, and axial tilt-are close enough to those of Earth to be acceptable and it is not too far from the Sun to be made inhabitable. In fact, recent computational studies utilizing climate models suggest that it could be possible to make Mars habitable with foresee125

126

•

Islands in the Sky

able technology. The essence of the situation is that while Mars' CO 2 atmosphere has only about 1% the pressure of the Earth's at sea level, it is believed that there are reserves of CO2 frozen in the south polar cap and adsorbed within the soil sufficient to thicken the atmosphere to the point where its pressure would be about 30% that of Earth. The way to get this gas to emerge is to heat the planet. In fact, the warming and cooling of Mars that occurs each Martian year as the planet cycles between its nearest and furthest positions from the Sun, in its slightly elliptical orbit, cause the atmospheric pressure on Mars to vary by plus or minus 25% compared to its average value on a seasonal basis. We cannot, of course, move Mars to a warmer orbit. However, we do know another way to heat a planet, through an artificially induced greenhouse effect that traps the Sun's heat within the atmo· sphere. Such an atmospheric greenhouse could be created on Mars in at least three different ways. One way would be to set up factories on Mars to produce very powerful artificial greenhouse gasses such as halocarbons (CFCs) and release them into the atmosphere. Another way would be to use orbital mirrors or other large-scale power sources to warm selected areas of the planet, such as the south polar cap, to release large reservoirs of the native greenhouse gas, CO2, which may be trapped there in frozen or adsorbed form. Finally natural greenhouse gasses more powerful than CO 2 ( but much less so than halocarbons) such as ammonia or methane could be imported to Mars in large quantities if asteroidal objects rich with such volatiles in frozen form should prove to exist in the outer solar system. Each of these methods of planetary warming would be enhanced by large amounts of CO 2 from the polar cap and from the soil that would be released as a result of the induced temperature rise. This CO2 would add massively to the greenhouse effect being created directly, speeding and multiplying the warming process. The Mars atmosphere/regolith greenhouse effect system is thus one with a built-in positive feedback. The warmer it gets, the thicker the atmosphere becomes; and the thicker the atmosphere becomes the warmer it gets. In the following sections, we'll show how this system can be modeled, and present the results of our calculations using such a model. Our results give very strong support to the belief that humans will. be able to effect radical improvements in the habitability of the Martian environment in the course of the twenty· first century.

Terraforming Mars

•

127

EQUATIONS FOR MODELING THE MARTIAN SYSTEM An equation for estimating the mean temperature on the surface of Mars as a function of the CO 2 atmospheric pressure and the solar constant is given by McKay and Oavis 1 as: ( 1) where T mean is the mean planetary temperature in Kelvin, S is the amount of solar output, where that of the present day Sun = 1, and P is the atmospheric pressure at Mars's mean surface elevation, given in bars. ( One bar is what flatlanders believe is normal atmospheric pressure, 14. 7 psi. Since people living in the fetid swamps near such major capitals as Washington, London, and Paris are influential in such things, this bizarre unit has become a standard.) Since the atmosphere is an effective means of heat transport from the equator to the pole, we estimate: T role

= Tmean

-

75(S

0•25)/( 1

+ SP)

(2)

We further assume, based upon a rough approximation to ob, served data, that: T max

= Tequator = 1.1 T mean

(3)

and that the global temperature distribution is given by:

(4) where 0 is the latitude (north or south). Equations (1) through ( 4) give the temperature on Mars as a function of CO 2 pressure. However, as mentioned above, the CO2 pressure on Mars is itself a function of the temperature. There are three reservoirs of CO 2 on Mars: the atmosphere, dry ice in the polar caps, and gas adsorbed in the soil. The interaction of the polar cap reservoirs with the atmosphere is well understood and is given simply by the relationship between the vapor pressure of CO 2 and the

128 •

Islands in the Sky

temperature at the poles. This is given by the vapor pressure curve for CO 2, which is approximated by:

P = 1.23 X 107{exp( - 3170{f pole)}

(5)

So long as there is CO 2 in both the atmosphere and the cap, equation (5) gives an exact answer to what the CO 2 atmospheric pressure will be as a function of polar temperature. However, if the polar temperature should rise to a point where the vapor pressure is much greater than that which can be produced by the mass in the cap reservoir (between 50 and 100 mb), then the cap will disappear and the atmosphere will be regulated by the soil reservoir. The relationship between the soil reservoir, the atmosphere, and the temperature is not known with precision. An educated guess is given in parametric form in reference 1 as: (6) where Ma is the amount of gas (in bars) adsorbed in the soil, C is a constant set so that equation (6) will reflect known Martian conditions, and Td is the characteristic energy required for release of gas from the soil (the "desorption temperature"). Equation (6) is essentially a variation of a well-known law for the change in chemical equilibrium with temperature, and so there is fair confidence that its general form is correct. However, the value of TJ is unknown and probably will remain so until after human exploration of Mars. While we don't know what the right value for TJ is, we can bracket the problem by varying Td from 15 to 40 K (the lower the value of Td the easier things are for prospective terraformers). Then we use the global temperature distribution given by equation ( 4) to integrate equation (6) over the surface of the planet to give us a global "soil pressure." This gives a reasonably accurate quasi 2-dimensional view of the atmosphere/regolith equilibrium problem in which most of the adsorbed CO2 is distributed to the planet's colder regions. Thus, in our model, regional ( in the sense of latitude) temperature changes, especially in the near-polar regions, can have as important a bearing on the atmosphere/regolith interaction as changes in the planet's mean temperature.

Terraforming Mars

•

129

RESULTS OF CALCULATIONS In Figure 8.1 we see the results of our model when applied to the situation at Mars' south polar cap, where it is believed that enough CO2 may be held frozen as dry ice to give Mars an atmosphere on the order of 50 to 100 mbar. We have plotted the polar temperature as a function of the pressure, in accord with equations (1) and (2), and the vapor pressure as a function of the polar temperature, in accord with equation (5). There are two equilibrium points, labeled A and B, where the values of P and T are mutually consistent. However, A is a stable equilibrium, while B is unstable. This can be seen by examining the dynamics of the system wherever the two curves do not coincide. Whenever the temperature curve lies above the vapor pressure curve, the system will move to the right, that is, toward increased temperature and pressure; this would represent a runaway greenhouse effect. Whenever the pressure curve lies above the temperature curve, the system will move to the left, that is, temperatures and pressure will both drop in a runaway icebox effect. Mars today is at point A, with 6 mbar of pressure and a temperature of about 147 K at the pole.

Effect of Martian Polar CO2

Greenhouse 220

-~

C1>

200

~

Vapor Pressure

•

Temp of Pole

0

a.

cu

C1> i... ::::::,

ci...u

180 160

C1>

C.

E

~

140 120 .1

10 1 Pressure (mb)

100

1000

Figure 8. I. Mars polar cap/atmosphere dynamics. Current equilibrium is at point A. Raising polar temperatures by 4 K would drive equilibria A and B together, causing runaway heating that would lead to the elimination of the cap.

130

•

Islands in the Sky

Now consider what would happen if someone artificially in, creased the temperature of the Martian pole by several degrees K. As the temperature is increased, points A and B would move toward each other until they met. If the temperature increase were 4 K, the temperature curve would be moved upward on the graph sufficiently so that it would lie above the vapor pressure curve everywhere. The result would be a runaway greenhouse effect that would cause the entire pole to evaporate, perhaps in less than a decade. Once the pressure and temperature have moved past the current location point B, Mars will be in a runaway greenhouse condition even without artificial heating, so if later the heating activity were discontinued the atmosphere would remain in place. As the polar cap evaporates, the dynamics of the greenhouse effect caused by the reserves of CO 2 held in the Martian soil come into play. These reserves exist primarily in the high,laritude regions, and by themselves are estimated to be enough to give Mars a 400,mbar atmosphere. We can't get them all out, however, because as they are forced out of the ground by warming, the soil becomes an increasingly effective "dry sponge" acting to hold them back. The dynamics of this system are shown in Figure 8.2, in which we

Greenhouse Effect of Martian Regolith 280

-... -...

:::ii:: (1) ~

cu (1)

•

Temperature of Tropic Summer

260 240 220 200

C.

E

~

180

~ Temp

---+-

160 .1

1

of Regolith Regolith Pressure

10

100

1000

Pressure (mb) Figure 8.2. Mars regolith/atmosphere dynamics under conditions of Td tile inventory of 500 mb of CO 2.

= 20 with

a vola-

Terraforming Mars

•

131

assume T.i = 20, current polar reserves of 100 mb, and regolith reserves of 394 mb, and graph the pressure on the planet as a function of the regolith temperature, Trcg,where Trci:is the average of the planet's temperature with different regions weighted in accordance With how much adsorbed gas they can hold at their own local temperature. Since Tregis a function of the temperature distribution and T mc,an, it is ultimately just a function of the pressure, and thus Trcg(P)can also be graphed. The result is a set of T(P) curves and P(T) curves, whose crossing points reflect stable or unstable equilibrium, just as in the case of the polar cap analysis. It can be seen in Figure 8.2 that the atmosphere soil system under the chosen assumption of TJ = 20 K has only 1 equilibrium point, which is stable, and which will be overrun by the pressure generated by the vaporized polar cap. Thus, by the time the process is brought to a halt, an atmosphere with a total pressure of about 300 mbar, or 4.4 psi, can be brought into being. Also shown in Figure 8.2 is the day-night average temperature that will result in Mars's tropical regions (T max) during summertime. It can be seen that the 2 73 K freezing point of water will be approached. With the addition of modest ongoing artificial greenhouse efforts, it can be exceeded. The assumption of TJ = 20 is optimistic, however, and the location of the equilibrium convergence point (point C in Figure 8.2) is very sensitive to the value chosen for TJ. In Figure 8.3 we show what happens if values of TJ = 25 and Td = 30 are assumed. In these cases, the convergence point moves from 300 mb at TJ = 20 to 31 and 16 mb for TJ = 25 and TJ = 30, respectively. (The value of the Tr,g curve in Figure 8.3 was calculated under the assumption of TJ = 25; it varies from this value by a degree or two for Td = 20 or 30.) Such extraordinary sensitivity of the final condition to the unknown value of TJ may appear at first glance to put the entire viability of the terraforming concept at risk. However, in Figure 8.3 we also show (dotted line) the situation if artificial greenhouse methods are employed to maintain Tregat a temperature 10 K above those produced by the CO 2 outgassing itself. It can be seen that drastic improvements in the final T and P values are effected for the Td = 25 and 30 cases, with all three cases converging upon final

132

Islands in the Sky

•

Atmosphere/Regolith Equilibria for Various

Td

260 -----·· Treg + 10 _rreg

240

_p,Td:20 -P,Td:25 --p,Td:30

~

.. 220

-

.c

------------

0 200 C>

... ....180 G)

160 1

10

100

1000

Pressure, mb Figure 8.3. An induced 10 Krise in regolith temperature can counter the effect of Td variations. Data shown assumes a planetary volatile inventory of 500 mb CO2.

states with Mars possessing atmospheres with several hundred mb of pressure. In Figures 8.4-8. 7 we show the convergence condition pressure and maximum seasonal average temperature in the Martian tropics

Equilibrium CO2Pressure on Poor Mars 5oo"T"'""---------------. 400

~ 300

ai

:i

200

f l:l.

100

en en

L--..-"'T"""--,..-~=:::;;:::::~=;:::i~~~ 15 20 25 30

35

-Pressure.

DT:20

-.-Pressure, -Pressure.

DT:10

__,.Pressure,

DT:O

DT:5

40

Regolith Gas Release Energy, Td Figure 8.4. Equilibrium pressure reached on Mars with a planetary volatile inventory of 500 mb CO2 after 50 mb polar cap has been evaporated. OT (aT in text) is artificially imposed sustained temperature rise.

T erraforming Mars

•

133

Maximum Tropical Temperature of Poor Mars ::.::

290

a; ... 280 :,

_T_

iii ... 270

.. DT:20

QI

Q.

E 260

~

-r_.,DT:10

250 cu u ·a 240

... .... ... 230

-----~--+-T_.,

0

QI

E E :,

DT:5

a-------------e

15

20

25

_T_,,

30

40

35

Regolith Gas Release Energy,

Cl)

DT:O

rd

Figure 8.5. Equilibrium maximum seasonal (diurnal average) temperature reached on Mars w1th a planetary volaule mventory of 500 mb CO 2 after 50 mb polar cap has been evaporated.

resulting in either a "poor" Mars, possessing a total supply of 500 mb of CO 2 (50 mb of CO 2 in the polar cap and 444 rnb in the regolith), or a "rich" Mars possessing 1000 rnb of CO 2 (100 mb in the polar cap and 894 mb in the regolith). Different

Equilibrium

CO 2 Pressure for Rich Mars

1000 ...---------------, 800 .c

E 600

... ::::, 400 ... 200 Cl) II) II)

Cl)

Q.

0 15

20

25

30

35

40

Regolith Gas Release Energy,

rd

Figure 8.6. Equilibrium pressure reached on Mars with a planetary volatile inventory of 1000 mb CO 2 after l 00 mb polar cap has been evaporated.

134

•

Islands in the Sky

Rich Mars Maximum Tropical Temperature 300 290 :::, c... a 280 Q) 0. 270 E 260 ~ ca 250 u ·a. 240 0 230

~

... Q)

-

.= E :::, E )(

......,__

......-

t>-------o-

15

20

25

30

--

35

Regolith Gas Release Energy,

T~.,.DT:5 T_,, DT:O

40 Td

ca

~

Figure 8.7. Equilibrium maximum seasonal temperature (diurnal average) reached on

Mars with a planetary volatile inventory of 1000 mb CO 2 after 100 mb polar cap has been evaporated.

curves are shown under the assumptions that either no sustained greenhouse effort is mounted after the initial polar cap release, or that continued efforts are employed to maintain the planet's mean temperature at a ~T of 5, 10, or 20° above the value produced by the CO 2 atmosphere alone. It can be seen that if a sustained effort is mounted to keep an artificial ~T of 20° in place, then a tangible atmosphere and acceptable pressures can be produced even if T J has a pessimistic value of 40 K. The important conclusion to be drawn from this analysis is that while the final conditions on a terraformed Mars may be highly sensitive to the currently unknown value of the regolith's outgassing energy, Td, they are even more sensitive to the level of sustained artificially induced greenhousing, ~T. Put simply, the final conditions of the atmosphere/regolith system on a terraformed Mars are control,

lable. Once significant regions of Mars rise above the freezing point of water on at least a seasonal basis, the large amounts of water frozen into the soil as permafrost would begin to melt, and eventually flow out into the dry riverbeds of Mars. Water vapor is also a very effective greenhouse gas, and since the vapor pressure of water on Mars would

Terraforming Mars

•

135

rise enormously under such circumstances, the reappearance of liquid water on the Martian surface would add to the avalanche of selfaccelerating effects all contributing to the rapid warming of Mars. The seasonal availability of liquid water is also the key factor in allowing the establishment of natural ecosystems on the surface of Mars. The dynamics of the regolith gas-release process are only approximately understood, and the total available reserves of CO 2 won't be known until human explorers journey to Mars to make a detailed assessment, so these results must be regarded as approximate and uncertain. Nevertheless, it is clear that the positive feedback generated by the Martian CO 2 greenhouse system greatly reduces the amount of engineering effort that would otherwise be required to transform the red planet. In fact, since the amount of a greenhouse gas needed to heat a planet is roughly proportional to the square of the temperature change required, driving Mars into a runaway greenhouse with an artificial 5 K temperature rise only requires about 1% the engineering effort that would be needed if the entire 50 K rise had to be engineered by brute force. The question we now examine is how such a 5 K global temperature rise could be induced.

METHODS OF ACCOMPLISHING GLOBAL WARMING ON MARS The three most promising options for inducing the required temperature rise to produce a runaway greenhouse·on Mars appear to be the use of orbital mirrors to change the heat balance of the south polar cap (thereby causing its CO 2 reservoir to vaporize), the importation of ammonia-rich objects from the outer solar system, and the production of artificial halocarbon (CFC) gases on the Martian surface. We'll look at each of these in tum. It may be, however, that synergistic 2 combination of several such methods may yield better results than any one of them used alone.

ORBITING MIRRORS While the production of a space-based sunlight-reflecting device capable of warming the entire surface of Mars to terrestrial tempera-

136

•

Islands in the Sky

tures is theoretically possible,3 the engineering challenges involved in such a task place such a project well outside the technological horizon of this article. A much more practical idea would be to construct a more modest mirror capable of warming a limited area of Mars by a few degrees. As shown by the data in Figure 8.1, a 5°K temperature rise imposed at the pole should be sufficient to cause the evaporation of the CO 2 reservoir in the south polar cap. Based on the total amount of solar energy required to raise the black-body temperature of a given area a certain number of degrees above the polar value of 150 K, it turns out that a space-based mirror with a radius of 125 km could reflect enough sunlight to raise the entire area south of 70° south latitude by 5 K. If made of solar sail-type aluminized Mylar material with a density of 4 tonnes/km 2, such a sail would have a mass of 200,000 tonnes. Many ships of this size are currently sailing the Earth's oceans. Thus, while this is too large to consider launching from Earth, if space-based manufacturing techniques are available its construction in space out of asteroidal or Martian moon material is a serious option. The total amount of energy required to process the materials for such a reflector would be about 120 MWe-years, which could be readily provided by a set of 5-MWe nuclear reactors such as are now being considered for use in piloted nuclear electric spacecraft. Interestingly, if stationed near Mars, such a device would not have to orbit the planet. Rather, solar light pressure could be made to balance the planet's gravity, allowing it to hover as a "statite" 4 with its power output trained constantly at the polar region. For the sail density assumed, the required operating altitude would be 214,000 km. The statite reflector concept and the required mirror size to produce a given polar temperature rise is shown in Figures 8.8 and 8.9. If the value of Td is lower than 20 K, then the release of the polar CO2 reserves by themselves could be enough to trigger the

Figure 8.8. Solar sails of 4 tonnes/km 2 density can be held stationary above Mars by light pressure at an altitude of 214,000 km. Wasting a small amount of light allows shadowing to be avoided.

Terraforming Mars

•

137

Heating Martian Pole with Mirrors 1000...---------------, Cl) Cl)

800

ur :,

600

cu :E

------

Mirror Radius (km) Mirror Mass (kilotonnes)

"C

cu a:

... 0

400

:i

... ...

200 0 0

5 Temperature

10

15

20

increase at Pole (K)

Figure 8.9. Solar sail mirrors w1th radii on the order of 100 km and masses of 200,000 tonnes can produce the 5 K temperature rise required to vaporize the C0 1 in Mars' south polar cap. It may be possible to construct such mirrors in space.

release of the regolith's reserves in a runaway greenhouse effect. If however, as seems probable, TJ is greater than 20 K, then either the importation or production of strong greenhouse gases will be required to force a global temperature rise sufficient to create a tangible atmospheric pressure on Mars.

MOVING AMMONIA ASTEROIDS Ammonia is a powerful greenhouse gas, and it is possible that nature has stockpiled large amounts of it in frozen form on asteroidal-sized objects orbiting in the outer solar system. If moving material from such objects to Mars is envisioned, then such orbits would be quite convenient, because strange as it may seem, it is easier to move an asteroid from the outer solar system to Mars than it is to do so from the Main Belt or any other inner solar system orbit. This odd result follows from the laws of orbital mechanics, which cause an object farther away from the Sun to orbit it slower than one that is closer in. Because an object in the outer solar system moves slower, it

138

•

Islands in the Sky

takes a smaller AV to change its orbit from a circular to an ellipse. Furthermore, the orbit does not have to be so ellirtical that it stretches from Mars to the outer solar system; it is sufficient to Jistort the object's orbit so that it intersects the path of a major rlanet, after which a gravity assist can do the rest. The results are shown in Figure 8.10. It can be seen that moving an asteroid rnsitioned in a circular orbit at 25 AU, by way of a Uranus gravity assist to Mars, requires a AV of only 0.3 km/s, compared to a 3.0 km/s .i V to move an asteroid directly to Mars from a 2. 7 AU position in the Main Belt. The time of flight required for such transfers is shown in Figure 8.11. Now we don't know for sure if there are numerous asteroid-si:e objects in the outer solar system, but there is no reason to helieve that there aren't. As of this writing, only one is knlnvn, hut that one, Chiron, orbiting between Saturn anJ Uranus, is rather large (180 km diameter), and it may be expected that a lot of small ohjccts can be found for every big one. In all probability, the outer solar system contains thousands of asteroids that we have yet to Jiscover because they shine so dimly compared to those in the Main Belt (the brightness of an asteroid as seen from Earth is inversely propor-

Velocity Change Required to Transport Asteroids to Mars (Gravity Assist at Intermediate Planets)

e-

5---------------

VI

4

~ "C

Cl> ...

3

er

2

:::s

Cl>

• ----

---------..&.

--+-

~------.

a:

-

-------

I

cu

'ii

0 ........

0

1.- ____

10

...;r..__

20

Jupiter

-----saturn

> C

Mars

--+Uranus ---Neptune

..a:::,.;~----'

30

40

50

Initial Distance of Asteroid from Sun (AU) Figure 8.10. Using gravity assists, the tl.V required to propel an outer solar system aster· oid onto a collision course with Mars can be less than 0.5 km/s. Such "falling" objects can release much more energy upon impact than is required to set them in motion.

T erraforming Mars

•

139

Asteroid Flight Time to Mars

(Gravity Assist Maneuver at Intermediate Planets)

200

en ... n,

150

Q)

~ Q)

E i=

-

.J:

100 ~"_..,.

~~----

50

--+--

Neptune

-------__

Uranus Saturn Jupiter Mars

_g,

u.

0

50 Initial Asteroid Distance from Sun (AU) 0

10

20

30

40

Figure 8.11. lhlli,rt, tl1~ht tinll's from the outer solar system to Mars are typically between ~ 'i .mJ 'iO ,·car,.

tional to the fourth power of its distance from the Sun). Furthermore, because water, ammonia, and other volatiles freeze so completely in the outer solar system, it is likely that the asteroids to be found beyond Saturn are largely composed of frozen gases (such appears to be the case for Chiron). This makes it possible for us to move them. Consider an asteroid made of frozen ammonia with a mass of 10 billion tonnes orbiting the Sun at a distance of 12 AU. Such an object, if spherical, would have a diameter of about 2.6 km, and changing its orbit to intersect Saturn's ( where it could get a transMars gravity assist) would require a LlV of 0.3 km/s. If a quartet of 5000-MW nuclear thermal rocket (NTR) engines powered by either fission or fusion were used to heat some of its ammonia up to 2,200 K (5000 MW fission NTRs operating at 2500 K were tested in the 1960s), they would produce an exhaust velocity of 4 km/s, which would allow them to move the asteroid onto its required course using only 8% of its material as propellant. Ten years of steady thrusting would be required, followed by about a 20-year coast to impact. When the object hits Mars, the energy released would be about 10 TW-years, enough to melt 1 trillion tonnes of water (a lake 140 km on a side and 50 meters deep). In addition, the ammonia released by a single such object would raise the planet's temperature by about 3 K and form a shield that would effectively mask the

140

•

Islands in the Sky

planet's surface from ultraviolet radiation. As further missions proceeded, the planet's temperature could be increased globally in accord with the data shown in Figure 8.12. Forty such missions would double the nitrogen content of Mars' atmosphere by direct importation, and could produce much more if some of the asteroids were targeted to hit beds of nitrates, which they would volatilize into nitrogen and oxygen upon impact. If one such mission were launched per year, within half a century or so most of Mars would have a temperate climate, and enough water would have been melted to cover a quarter of the planet with a layer of water 1 m deep. While attractive in a number of respects, the feasibility of the asteroidal impact concept is uncertain because of the lack of data on outer solar system ammonia objects. Moreover, if T 1 is greater than 20 K, a sustained greenhousing effort will be required. As the characteristic lifetime of an ammonia molecule on Mars is likely to be less than a century, this means that even after the temperature is raised, ammonia objects would need to continue to be imported to Mars, albeit at a reduced rate. As each object will hit Mars with an energy yield of about 70,000 I-megaton hydrogen bombs, the continuation of such a program may be incompatible with the objective of making Mars suitable for human settlement.

Heating Mars with Imported Ammonia f::,

1000-.------------------, NH, Pressure (microbars)

Cl) Cl)

~

0.

100

'X."

z

0

C

0

10

·u;

-

#NH, Missions (10 Gt each)

.!!

:::E

1---...-------------0 10

20

30

40

Global Temperature Increase (K)

Figure 8.12. Importing four IO-billion-tonne ammonia asteroids to Mars would impose an 8 K temperature rise, which after amplification by CO 2 feedback could create drastic changes in global conditions.

Terraforming Mars

•

141

A possible improvement to the ammonia impact method is suggested by ideas given in reference 5, where it is pointed out that bacteria exist which can metabolize nitrogen and water to produce ammonia. If an initial greenhouse condition were to be created by ammonia object importation, it may be possible that a bacterial ecology could he set up on the planet's surface that would recycle the nitrogen resulting from ammonia photolysis back into the atmosphere as ammonia, thereby maintaining the system without the need for further impacts. Similar schemes might also be feasible for recycling methane, another short-lived natural greenhouse gas which might be imported to the planet.

PRODUCING HALOCARBONS ON MARS In Table 8. l we show the amount of halocarbon gases (CFCs) needed in Mars' atmosphere to create a given temperature rise, and the power that would be needed on the Martian surface to produce the required CFCs over a period of 20 years. If the gases have an atmospheric lifetime of 100 years, then approximately one-fifth the power levels shown in the table will be needed to maintain the CFC concentration after it has been built up. For purposes of comparison, a typical nuclear power plant used on Earth today has a power output of about 1000 MWe and provides enough energy for a medium-sized (Denver) American city. The industrial effort associated with such a power level would be substantial, producing about a trainload of refined material every day and requiring the support of a work crew of several thousand people on the Martian surface. A total project Table 8. 1. Greenhousing Mars with CFCs

Induced Heating (K)

CFC Pressure (micro-bar)

CFC Production (tonnes/hour)

Power Required (MWe)

5 10 20 30 40

0.012 0.04 0.11 0.22 0.80

263 878 2414 4829 17569

1315 4490 12070 24145 87845

142

•

Islands in the Sky

budget of several hundred billion dollars might well be required. Nevertheless, all things considered, such an operation is hardly likely to be beyond the capabilities of the mid-twenty-first century. Using such an approach, Mars could be transformed from its current dry and frozen state into a warm and slightly moist planet capable of supporting life for several decades. Humans could not breathe the air of the thus transformed Mars, but they would no longer require space suits and instead could travel freely in the open wearing ordinary clothes and simple SCUBA-type breathing gear. However, because the outside atmospheric pressure will have been raised to human-tolerable levels, it will be possible to have large habitable areas for humans under huge domelike inflatable tents containing breathable air. In contrast, simple hardy plants could thrive in the CO 2-rich outside environment, and spread rapidly across the planet's surface. In the course of centuries, these plants would introduce oxygen into Mars' atmosphere in increasingly breathable quantities, opening up the surface to advanced plants and increasing numbers of animal types. As this occurred, the CO2 content of the atmosphere would be reduced, which would cause the planet to cool unless artificial greenhouse gases were introduced capable of blocking off those sections of the infrared spectrum previously protected by CO 2. The halocarbon gases employed would also have to be varieties lacking in chlorine, if an ultraviolet shielding ozone layer is to be built up. One good candidate for such a gas would be perfluoromethane, CF 4, which also has the desirable feature of being very long-lived (about 10,000 years) in an upper atmosphere. Providing these matters are attended to, however, the day would eventually come when the domed tents would no longer be necessary.

ACTIVATING THE HYDROSPHERE The first steps required in the terraforming of Mars, warming the planet and thickening its atmosphere, can be accomplished with surprisingly modest means using in situ production of halocarbon gases. However, the oxygen and nitrogen levels in the atmosphere would be too low for many plants, and if left in this condition the planet would remain relatively dry, as the warmer temperatures took centuries to melt Mars' ice and deeply buried permafrost. It is in

Terraforming Mars

•

143

this, the second phase of terraforming Mars, during which the hydrosphere is activated, the atmosphere made breathable for advanced plants and primitive animals, and the temperature increased further, that either space-based manufacturing of large solar concentrators or human activity in the outer solar system is likely to assume an important role. Activating the Martian hydrosphere in a timely fashion will require doing some violence to the planet, and, as discussed previously, one way this can be done is with targeted asteroidal impacts. Each such impact releases the energy equivalent of 10 TW-yrs. If Plowshare methods of shock treatment for Mars are desired, then the use of such projectiles is certainly to be preferred to the alternative option 3 of detonation of hundreds of thousands of thermonuclear explosives. After all, even if so much explosive could be manufactured, its use would leave the planet unacceptably radioactive. The m,e of orbiting mirrors provides an alternative method for hydrosphere activation. For example, if the 125-km radius reflector discussed earlier for use in vaporizing the pole were to concentrate its power on a smaller region, 27 TW would be available to melt lakes or volatilize nitrate beds. This is triple the power available from the impact of a IO-billion tonne asteroid every year, and in all probability would be far more controllable. A single such mirror could drive vast amounts of water out of the permafrost and into the nascent Martian ecosystem very quickly. Thus, while the engineering of such mirrors may be somewhat grandiose, the benefits to terraforming, of being able to wield tens of TW of power in a controllable way can hardly be overstated.

OXYGEN A TING THE PLANET The most technologically challenging aspect of terraforming Mars will be the creation of sufficient oxygen in the planet's atmosphere to support animal life. While primitive plants can survive in an atmosphere without oxygen, advanced plants require about 1 mb and humans need 120 mb. While Mars may have super-oxides in its soil or nitrates that can be pyrolysed to release oxygen (and nitrogen) gas, the problem is the amount of energy needed: about 2200 TW-yrs for every mb produced. Similar amounts of energy are

144 •

Islands in the Sky

required for plants to release oxygen from COz. Plants, however, offer the advantage that once established they can propagate them, selves. The production of an oxygen atmosphere on Mars thus breaks down into two phases. In the first phase, brute force engineering techniques are employed to produce sufficient oxygen (about 1 mb) to allow advanced plants to propagate across Mars. Assuming there are three 125,km radius space mirrors active in supporting such a program and sufficient supplies of suitable target material on the ground, such a goal could be achieved in about 25 years. At that point, with a temperate climate, a thickened CO2 atmosphere to supply pressure and greatly reduce the space radiation dose, and a good deal of water in circulation, plants that have been genetically engineered to tolerate Martian soils and to perform photosynthesis at high efficiency could be released together with their bacterial symbiotes. Assuming that global coverage could be achieved in a few decades and that such plants could be engineered to be 1% efficient (rather high, but not unheard of among terrestrial plants), then they would represent an equivalent oxygen-producing power source of about 200 TW. By combining the efforts of such biological systems with perhaps 90 TW of space-based reflectors and 10 TW of installed power on the surface (terrestrial civilization today uses about 12 TW), the required 120 mb of oxygen needed to support humans and other advanced animals in the open could be produced in about 900 years. If more powerful artificial energy sources or still more efficient plants were engineered, then this schedule could be accelerated accordingly, a fact which may well prove a driver in bringing such technologies into being. It may be noted chat thermo, nuclear fusion power on the scale required for the acceleration of terraforming also represents the key technology for enabling piloted interstellar flight. If terraforming Mars were to produce such a spinoff, then the ultimate result of the project will be to confer upon human, ity not only one new world for habitation, but myriads.

CONCLUSION What we have shown in this article is that within broad tolerances of uncertainty of Martian conditions, that drastic improvements in the life,sustaining characteristics of the environment of the red

T erraforming Mars

•

145

planet may be effected by humans using early to mid-twenty-first century technologies. While our immediate descendants cannot expect co use such near-term methods to "terraform" the planet in the full sense of the word, it at least should be possible to rejuvenate Mars, making it again as receptive to life as it once was. Moreover, in the process of modifying Mars, they are certain to learn much more about how planets really function and evolve, enough perhaps to assure wise management of our native planet. Beyond such near-term milestones, the tasks associated with full terraforming become more daunting and the technologies required more speculative. Yet, who can doubt chat if the first steps are taken, that the developments required to complete the job will not follow, for what is ultimately at stake is an infinite universe of habitable worlds. Seen in such light the task facing our generation-that of exploring Mars and learning enough about the planet and the methods of unlizing its resources to begin to transform it into a habitable planet-could not be more urgent, or more noble.

Chapter 9 Martyn

J. Fogg

A Planet Dweller's Dreams

Many of you reading this article would like to believe that the stars are our ultimate destination. One day, a branch of our species, taking with it other life forms from Earth, may embark upon the grandest odyssey of all: a million-year galactic diaspora, traveling from star to star, exploring, discovering, and learning, gradually dispersing throughout the spiral arms and disc of the Milky Way. It is possible to imagine an almost infinite variety of settings within which our space-faring descendants might choose to settle. However, what will the overall picture be like, where will our homes of the future be? If the imaginations of science fiction authors are to be a reliable guide, the answer is other planets about other stars. Extra-solar planets remain a major theme within SF. The majority of future histories, whether composed of one novel or many, see planets as being the fundamental building blocks of interstellar empires. Yet, this has not been taken for granted by a large fraction of space scientists who speculate over interstellar colonization. Especially from the mid-1970s onward, after Gerard O'Neill's pioneering work on space colonies, the emphasis of many has switched to the colonizationof space itself. By this I mean the construction of large, 147

148 •

Islands in the Sky

free-floating, artificial habitats, enclosing an environment comfortable for human beings. One can think of these space colonies as being extraterrestrial city-states or daughter biospheres of the Earth-miniaturized maybe, but still large enough to support a population of thousands of people. O'Neill's cohorts coined the term "planetary chauvinism" to attack what they saw as an outdated view that planets are the natural goal of any future space colonization effort. Their reasoning is quite persuasive: Space is rich in solar energy, which is free; space is rich in easily accessible raw materials, asteroids, comets, and the like, which are also free; and planets share an unfortunate property called a gravity well which makes travel to and fro much more expensive. An additional ingredient, when looking at the interstellar perspective, is that theoretical studies of the likely abundance of Earth-like planets about other stars indicate that they may be very rare. The most optimistic study to date was that of Stephen Dole of the Rand Corporation, who published a report in 1964 which estimated that about 1 in 200 stars possess a habitable planet. This means that we would have to search, on average, about 25 lightyears of space to locate such a world. In contrast to this, William Pollard of the ORAU Institute for Energy Analysis estimated that only 1 in 100,000 to 1 in 10,000,000 stars may be encircled by an Earth-like planet, requiring a search for a needle in a haystack 200 to 1,000 light-years in radius! Thus, argued the space colony enthusiasts, planet dwelling as a life-style will be outgrown by the human species. Instead, large colony-starships known as "worldships,11 such as Arthur C. Clarke's Rama, will travel the space lanes setting up shop in any convenient star system (for almost all are likely to possess asteroidal material} where a new, entirely spacebased civilization can be established. As I said, their reasoning is very persuasive, but as many SF authors have implicitly recognized-and they do have to entertain us-it is rather boring! Personally I'd rather live long-term on the outer skin of a large ball of rock than on the inside of a smaller metal can. I confess therefore to being a reactionary planetary chauvinist with an interest in re-engineering those planets that don't suit into a condition where they do, a process called terraforming. Having conducted and published terraforming-related research for several years, I decided to look at the potential effect of terraforming

A Planet Dweller's Dreams

•

149

processes on interstellar colonization. Now obviously, a starship is not going to have the capability to terraform any terrestrial planet it finds and so, for the dreary logic of the space colony brigade to be overcome, there must exist a large extra-solar population of nearEarth-like worlds, planets that would be easy to terraform. Is this possible, or even likely? The study I completed and published in the Journal of the BritishInterplanetarySociety1 suggests the answer might be yes.

DEFINITIONS A problem with comparing past estimates of the prevalence of habitable planets is that the relevant studies on which they are based all adopt a different meaning of the word "habitable." So, before we go on any further, I'd better define what I mean. We're going to speak of three different types of terrestrial planet.

l. Habitable Planet (HP): A world with an environment sufficiently similar to the Earth as to ·allow comfortable and free human habitation. 2. BiocompatiblePlanet (BP): A planet possessing the necessary physical parameters for life to flourish on its surface. If initially lifeless, then such a world could host a biosphere of considerable complexity without the need for terraforming. 3. Easily TerraformablePlanet (ETP): A planet that might be rendered biocompatible, or possibly habitable, and maintained so by modest planetary engineering techniques and with the limited resources of a starship or robot precursor mission. Now, it is plain that these three planetary categories are nested: (1) being a subset of (2), being a subset of (3). Clearly the solar system at the present time contains just one ETP, which also happens to be a habitable planet, the Earth. However, this was not always so. The history of our solar system as revealed by planetary scientists suggests that early on, two qther planets could have been included within the groupings defined previously. With a sample of more than one, our ability to extrapolate to the likely situation about other

150 •

Islands in the Sky

stars is much improved. The best way to approach this problem is by studying and understanding the history of the Sun's ecosphereand the planets within it.

THE ECOSPHERE AND BIOCOMPATIBLE PLANETS One can define the ecosphere as that zone surrounding a star within which conditions are thermally compatible with life. It is a finite volume with inner and outer boundaries: too close to a star and a planet's water exists as vapor, too far away and it is permanently frozen. Since we assume a crucial requirement of life is the presence of stable liquid water on a planetary surface, then the ecosphere is the place where we might expect to find biocompatible planets. The boundaries of the ecosphere are obviously controlled primarily by the amount of radiation received from the central star. (The Earth receives about 1370 W /m 2 of solar energy, an illuminance known as the solar constant, which I shall denote by the letter S.) However, the luminosity of the star itself is not constant over long time scales but gradually increases throughout its main sequence lifetime. Our Sun was only about 70% as bright 4.6 billion years ago, when newly born, than it is now, and so the ecosphere is not a static volume but one which expands gradually outward. Planets can thus move in or out of the ecosphere as the central star evolves; moreover there are processes of planetary evolution which can result in a planet losing its liquid water so that, even though it may remain in the ecosphere, it ceases to be biocompatible. Therefore, the dimension of time, as well as distance, is of importance in our discussion. One problem is that we don't actually know the true extent of the Sun's ecosphere. As astronomical theory has advanced, estimates of the width of this biocompatible zone have fluctuated. Before the space age, it was thought that Mars and Venus might be biocompatible, implying that the ecosphere extended to their orbits and maybe beyond. By the mid-1970s, modeling of terrestrial evolution was suggesting that the ecosphere was wafer thin and the Earth only situated within it by the merest of flukes. The most modem research,

A Planet Dweller's Dreams

•

151

however, has swung the other way; it now seems that the ecosphere may extend farther from the Sun than hitherto thought. One of the conundrums that has dogged planetary scientists is the so-called "Faint Young Sun Paradox." As far back as we can look in the Earth's geological record, we find evidence for the presence of liquid water. This implies a global average surface temperature maintained consistently above freezing, despite the changing luminosity of the Sun. Yet, the young Earth, illuminated by sunlight with only 70% its present brightness (0.7 S), should have been frozen and should have stayed frozen up until comparatively recently. The fact that this didn't happen indicates that there must have been some mechanism at work capable of keeping the Earth's surface warm even though the planet was receiving less energy; a selfregulatory mechanism rather akin to that of a warm-blooded animal, which can automatically keep its body temperature constant whether the day is sunny or cool. In 1981 James Walker and coworkers at the University of Michigan proposed just such a mechanism, the Carbonate-Silicate Cycle, which is capable of linking the amount of sunlight received by a terrestrial planet with the greenhouse effect of its atmosphere. It hinges on the fact that carbon dioxide in the atmosphere of a planet with liquid water on its surface is not stable-it weathers silicate rocks, reacting with them chemically to produce carbonate minerals, such as calcite, the main constituent of limestone. Why then does the Earth's atmosphere contain CO 2? This is because carbonate minerals are not stable either, under conditions of high temperature and pressure. Since the Earth's crust is tectonically active ( it moves about, overturning parts of itself over time scales of millions of years), then carbonate rocks can be deeply buried where they become heated and squeezed. This drives the carbon dioxide out of the mineral phase whereupon it escapes back into the atmosphere via volcanoes. The clever bit is that the weathering rate at the surface is sensitive to temperature. If the planet cools then the weathering rate falls. However, the rate of volcanic outgassing of CO 2 stays the same because of the delays inherent in the cycle and so the amount of CO2 in the atmosphere increases. This strengthens the atmosphere's greenhouse effect, producing a warming trend to counteract the original cooling. Conversely, if the planet warms up for any reason, surface weathering increases, CO 2 is drawn down out of the atmo-

152

•

Islands in the Sky

sphere, reducing the greenhouse effect and producing a cooling. What we in fact have is a planetary engine that operates automatically to maintain temperatures a few tens of degrees above the freezing point of water-conditions ideally suited for life! The Faint Young Sun Paradox for the Earth may thus be explained. Surface temperatures have been maintained constant over our planet's history by a controlled decline of the atmospheric partial pressure of CO 2 from an initial value of about 1 bar to the 0.0003 bar of today-a change that would have kept pace with the brightening of the Sun. However, this is not all of the explanatory power of the carbonate-silicate cycle; it also enables us to explain the history of Mars and to estimate the position of the outer edge of the ecosphere. Mars was not always the frozen desert it is today. Images returned by space probes have revealed areas of the planet cut by what look like dried-up river beds; there are desiccated lake beds too, and some even claim to be able to point out the fossil shore lines of an empty ocean. These structures mostly occur on the most ancient of Martian terrain, that dating from the first billion years of the planet's history. Thus, it seems that Mars experienced an early warm and wet epoch, a time when water ran and pooled on its surface, a time when Mars was a biocompatible planet, irrespective of whether life actually evolved there. How can we reconcile a faint young Sun with a warm Mars? Well, if early Mars had an atmosphere of about 5 bars of carbon dioxide, with maybe other greenhouse gases too, then surface temperatures could have been at or a little above freezing. It is thought that this atmosphere could have been maintained by an active carbonate-silicate cycle, CO 2 being resupplied by intense volcanism powered by the internal heat of a young, recently formed, world. Why then is Mars still not like this? Why instead did the atmosphere dwindle, the seas dry up, and the planet's temperature plummet to -60 C? After all, the Sun is now hotter than it once was. The answer is that Mars is small, smaller and less massive than Earth and therefore has a larger surface-to-volume ratio. The little world lost its internal heat much more rapidly. Its crust thickened until it could no longer bury and decompose its carbonates. The carbonate-silicate cycle was cut. From that point on, biocompatible Mars was doomed. Its air reacted permanently with the surface; even after all of Mars' water froze, this process continued gradually to

A Planet Dweller's Dreams

•

153

leave the planet with the tenuous atmospheric remnant we see today. Mars died, even though it was situated within the ecosphere. Yes! The fact that Mars was once biocompatible suggests that the outer boundary extends beyond the Martian orbit. A planet at about 1.8 AU from the present,day Sun receives about as much radiation as a young Mars, about 0.3 S, and so we can speculate that any terrestrial planet with active volcanism and tectonics, capable of sustaining an active carbonate,silicate cycle, could remain biocompatible out to this dis, tance or a little farther. Now let's look closer co the Sun for the inner edge of the eco, sphere. Computer models of the Earth's climate suggest that were our planet's illuminance to be increased to 1.1 S, the oceans would start to evaporate. Large quantities of water vapor would start to accumulate in the atmosphere, a sufficient mass to keep a lid on a sort of planetary pressure cooker which would keep the oceans from disappearing entirely. Although water vapor is a powerful greenhouse gas, the increase in atmospheric pressure would prevent the boiling point of water, which itself increases with pressure, from being over, taken by rising surface temperatures. The Earth would be in a climatic regime known as the wet greenhouse with a 2,bar atmosphere of half nitrogen and half water vapor overlying scalding oceans at about 100 C. Increasing our planet's illuminance to 1.4 S would cause the pressure cooker to blow; the oceans would evaporate entirely, and the surface would dry out, reducing the ease with which volcanically produced carbon dioxide could weather back into the surface. Here we have a situation called the runaway greenhouse, a state in which we currently find the planet Venus. However, the Venusian atmo, sphere is now very dry. If, as seems likely, Venus once had a similar amount of water to the Earth, then where has it all gone? Both greenhouse states initially result in an atmosphere which is so wet that, unlike on Earth, large quantities of water vapor can rise to the edge of space. Here, ultraviolet radiation in sunlight can dissociate water molecules into hydrogen and oxygen. Hydrogen being a very light gas escapes to space; some of the oxygen does too, while the rest reacts with surface rocks. It has been calculated that an amount of water equivalent to an entire terrestrial ocean can be lost in this way in about 500 million years. If these conclusions are correct, then at about 1.1 S, equivalent to a distance from the Sun of 0.95 AU, the wet greenhouse becomes

154

•

Islands in the Sky

established-a state incompatible with life of almost any sort. We have reached the inner edge of the ecosphere. Since the Sun was significantly less luminous when it was young, might Venus have once been within the ecosphere? The answer is no: the minimum illuminance Venus might have received was about 1.38 S, so that it was gripped by a greenhouse climate from the start. However, it may have started off in the wet greenhouse before evolving a few hundred million years later to a full runaway. Thus, during the earliest history of the solar system, there may have been threeterrestrial planets with oceans, two of which were within the ecosphere and at least one of which gave rise to life.

EASILY TERRAFORMABLE AND HABIT ABLE PLANETS To complete our discussion of the three categories of planets we are interested in, we must briefly talk about ETPs and HPs. The former may be approached by asking the question, "Will it be easy to terraform Mars and Venus?" (Remember that the word "easy" in this context is defined as being within the wherewithal of a space colony/starship, the size of a city-state or small nation.) Well, the answer is a fairly firm no and if you have taken in the discussion in the last section it should be clear why. Mars has lost its ability to recycle CO 2, with the result that its ancient atmosphere has become literally petrified into the rocks of its surface. Recreating an atmosphere thick enough to make Mars biocompatible again would be a huge undertaking. It could be done, but not as previously thought by simply warming the planet with orbiting mirrors. The gases would need to be shocked out of the crust by buried explosions or impacts-replacing the internal heat that the planet has lost-or even imported from elsewhere in the solar system (see my article in the January 1991 Analog). This could perhaps be achieved by a mature civilization but not by the inhabitants of one ship lately arrived from a long interstellar journey. For Venus we will have to replace an ocean's worth of water, and the large mass of its CO 2 atmosphere must be eliminated. The old idea that all of this could be achieved by dumping a few photosyn-

A Planet Dweller's Dreams

•

155

thetic algae into the planet's cloud deck, waiting a few years and then "bingo! We have a habitable planet" is, of course, bunkum. Any realistic proposal for terraforming Venus (see the reference list) inevitably requires a massive project well beyond the capabilities of a few hundred thousand interstellar travelers. However, as we have seen, conditions on these two worlds may have been much less hostile during the first billion years of their respective histories. In fact, both planets at that time would have been much easier to terrafarm than at present. Mars was already biocompatible and could have been made more so by raising its illuminance. This would have increased the weathering rate at its surface,reducing the mass of carbon dioxide in the atmosphere. This process could have been speeded further by implanting photosynthetic life on Mars capable of using solar energy to use still more CO 2 to increase its own biomass, releasing oxygen in the process. While Venus was in the wet greenhouse stage, a reduction in its illuminance to below 1.1 S would have caused most of the water in the atmosphere to rain out, resulting in a relatively rapid transition from greenhouse to more Earth-like conditions. Now the modification of a planet's illuminance is likely to be one of the simplest and cheapest planetary engineering techniques. Construction of large orbiting sunshades or mirrors from hyperthin aluminum solar sail material would not be too difficult, especially to a civilization used to obtaining and processing space resources. This is the sort of terraforming that one can imagine interstellar travelers getting to grips with straight away. Thus, in the context of the model of the ecosphere discussed previously, we can define planets, theset of easily terrafarmableplanets to consistof all biocompatible plus planets in the wet greenhouse state that have retaineda substantial fraction of their water. Now we just have to put the icing on our definition of a habitable planet. Of course, the fact that the Earth is habitable is the one data point we are certain of; however, this has not always been the case. In fact, for most of the history of our planet, you or I could not have survived exposure to the environment unprotected. We would have suffocated, gasping for the 21% oxygen we are used to from air that contained less than 1%. It is thought that oxygen levels in the atmosphere only approached present levels about 600 million years ago at the beginning of the Cambrian period, when the Earth

156

•

Islands in the Sky

experienced a rapid and dramatic proliferation of complex lifeforms-the so-called "Cambrian Explosion." Before that stretches the long eons of the Precambrian age, the age of anaerobic lifebacteria, algae, and so forth-floating in the photic zones of the seas or living in the slime on the ocean floor. The Earth in the Precambrian age was not therefore a habitable planet, but was of course biocompatible. Such a world would be even easier to terraform than a young Mars or Venus. Since our model of the ecosphere is based on the carbonatesilicate cycle, which links the level of planetary illuminance with the chemical makeup of its atmosphere, we can take a leap of faith and choose the illuminance of the Earth at the beginning of the Cambrian age (0.94 S) as the threshold at which a planet with Earth-like parameters can be considered habitable. Thus, the set of habitableplanetsis assumed to be those biocompatibleplanets with an

illuminanceof > 0. 94 S.

WE NEED TO SUMMARIZE! If you don't know much about planetology, then some of you may be feeling a little punch-drunk by now. A summary of the model just described is definitely needed, something preferably which sums up the whole picture in one diagram. Have a look at Figure 9.1. On the vertical axis we have a scale of illuminance. All biocompatible planets are contained within the ecosphere that lies between 0.25 S and 1.1 S and is divided into three zones: Juvenile Martian, where planets could be similar to the early Mars; Juvenile Terran, where planets would be similar to the Precambrian Earth, and the Habitable,which we have already defined. The horizontal axis is a scale of planetary volcanic/tectonic activity which you needn't worry about other than to note that the central dividing line where the scale reads zero is the point at which planets cool to the extent that they cannot sustain a closed carbonate-silicate cycle. Since both solar luminosity and volcanic activity vary with time, the position of a planet on the diagram changes, tracing out an evolutionary track. Such tracks are shown on the diagram for the Earth, Mars, and Venus, each one being marked at intervals of 1 billion years.

•

A Planet Dweller's Dreams

157

AUTOMATIC

RUNAWAY

GREENHOUSE

1·5

1·3

\

I

,.,,• ;;,1-1r-----.._

\

\ I

\

I

___

\

I I

\ I I

\

I

I

\

I

\

I

\

.I

\ \

I

\

(

\

I

I \

I

\

I

\

I

\

I

\

I

\

1·2

\

I

\

I

\

I

I\

RUNAWAY EENHOUSE, ,

\ I

I

/

sT GREENHOUSE MOf ,, ____ __,, ...a,_ ____

I

\

I

\

I

\

I \

I

,.__ ___

I

\ \

\

\ \ \

\

\ \

~

_ _______ ~-:_________

I

•l! 1·0

Habitable

•

7

l o-9 ---------C

4

Juvenll• Terran

0·8

0 2 1

0·7

-------------·-----

0·6

Juvenll• Mertlen

0·5 0·4

0·31--------------:-, 0·2

0·1 ·1

·1

Figure 9. 1. Climatic evolutionary tracks for Venus, the Earth, and Mars, comparing illuminance versus geological activity over time, marked in intervals of 1 billion years on the respective curves. The central dividing line at log(T/T PT) = 0 represents the point in a planet's geological evolution at which it can no longer sustain plate tectonics or a geochemical carbon cycle.

Now, what does Figure 9.1 tell us? Early in the history of the solar system, both the Earth and Mars were within the ecosphere and were warm and wet. Venus may have been in the wet greenhouse region, with oceans on its surface under a sweltering water vapor

158

•

Islands in the Sky

atmosphere. Shortly afterward, the seas of Venus evaporated permanently, the loss of water to space from the upper atmosphere desiccating the planet for good. At 1 billion years or thereabouts, Mars leaves the ecosphere for the "runaway ice house," being no longer able to recycle its carbon dioxide atmosphere and to prevent the onset of a permanent ice age. Only the Earth remains biocompatible. After 4 billion years it becomes habitable, and 600 million years later we appear to ask the questions about how it all happened in the first place. Now we have a model robust enough to allow us to speculate in a reasonably sensible way concerning the situation about other stars. Figure 9.1 looks quite hopeful. After all, at one time the solar system contained three easily terraformable planets, instead of one. The habitable zone within the ecosphere, which is the approximate volume most other studies concentrate on, is seen to be just a small part of a much bigger potentiallyhabitable zone.

EXPLORING AN EMPIRE OF STARS In order to estimate the abundance of ETPs in our part of the galaxy, I embedded this model of the ecosphere within a broader astronomical framework. It became the centerpiece of a Monte Carlo computer simulation capable of generating random and unique planetary systems about stars of varying age and mass. The computer plays a kind of planet-building game, following a large number of steps and shaking dice each time. It starts with the creation of the star itself, followed by the formation of its planetary system and terminating with an evaluation of the climatic evolution of its terrestrial planets. The general strategy of the program is shown in Figure 9.2. The ultimate aim of the calculations is to determine the surface temperature range of a terrestrial planet to see if it qualifies as one of the three types of worlds, defined previously, which interstellar travelers might search for. The rules of the game cannot, of course, be based solely on rigorous theory as we have nowhere near a full understanding of planetology and have yet to detect an extra-solar planetary system. However, the laws of physics, empirical relations derived from observation, and current theory are blended together within the algorithms of the program such that the computer pro-

A Planet Dweller's Dreams

•

159

AGE

Tld ■ I

lnltl ■ I

Forces

Rotation

FINAL ROTATION

Planetary Mass

Stellar Man

I I I

I

!~~:!

M,L Rel•tlon

I I I I

GEOLOGICAL ACTIVITY

Stellar Luminosity

I

I I

: Geochemlcal 1Carbon Cycle

Dlat ■ nce

ol Planet

I I I I

MEAN ILLUMINANCE

ORBITAL ECCENTRICITY

AXIAL INCLINATION

Greenhou11 Effect

Surface T■ mper ■ ture R ■ nge

Figure 9.2. Stellar and planetary parameters that ultimately determine the range of planetary surface temperature. Dashed arrows represent speculative linkages.

duces physically realistic, "sensible," although not necessarily stereotypical, planetary systems (see Figure 9.3 ). I'm not going to cover the technical aspects of the simulation here; those of you who are interested and want more details can obtain it from the references listed at the end of this article. Suffice it to say that since the mass distribution and spatial density of stars

160

•

Islands in the Sky

-----------~ ·RO

•OIi

•03

•IIR

•26

801

e,a

•II•

't

•'11